High Yield Notes

Quantitative Methods 2023 Level II High Yield Notes
LM01 Basics of Multiple Regression and Underlying Assumptions

Uses of multiple linear regression
Multiple regression allows us to determine the effect of more than one independent
variable on a particular dependent variable.
Multiple regression can be used:
 To explain the relationships between financial variables: e.g., the relationship
between inflation, GDP growth rates and interest rates.
 To test existing theories – e.g., are equity returns impacted by a stock’s market cap
and value/growth factors.
 To make forecasts – e.g., using variables such as financial leverage, profitability,
revenue growth, and changes in market share to predict whether a company will face
financial distress.
The basics of multiple linear regression

A multiple linear regression model has the general form:
Yi = b0 + b1 X1i + b2 X2i + ⋯ + bk Xki + εi , i = 1, 2,….n
The slope coefficient bj measures how much the dependent variable Y changes when the
independent variable, Xj, changes by one unit holding all other independent variables
constant.
The intercept coefficient b0 represents the expected value of Y if all independent variables
are zero.
For example, consider the following regression equation:
Y = 0.2 + 0.6X1 + 0.5 X2 + ϵ
If X1 changes by 1 unit and X2 remains constant, then Y will change by 0.6 units. Similarly, if
X1 remains constant and X2 changes by 1 unit, then Y will change by 0.5 units. If X1 and X2
are each zero, then the expected value of Y is 0.2.
Assumptions underlying multiple linear regression

The five main assumptions underlying multiple regression models are:
1. Linearity: The relationship between the dependent variable and the independent
variables is linear.
2. Homoskedasticity: The variance of the regression residuals is the same for all
observations.
3. Independence of errors: The observations are independent of one another. This
© IFT. All rights reserved 1

implies the regression residuals are uncorrelated across observations.

4. Normality: The regression residuals are normally distributed.
5. Independence of independent variables:
5a. Independent variables are not random.
5b. There is no exact linear relation between two or more of the independent
variables or combinations of the independent variables.
Commonly used diagnostic plots
Scatterplots of dependent and independent variables are used to check if the assumptions
of ‘linearity’ and ‘independence of independent variables’ have been violated.
Scatterplots of residuals is used to check if the assumptions of ‘homoskedasticity’ and
‘independence of errors’ have been violated.
A ‘Q-Q’ plot of residuals is used to check if the assumption of ‘normality’ has been violated.

LM02 Evaluating Regression Model Fit and Interpreting Model Results

Goodness of fit
R2 and adjusted R2
R2 measures the percentage of variation in Y that is explained by the independent variables.
In multiple regression, R2 increases as we add new independent variables, even if the
̅2
amount of variation explained by them is not statistically significant. Hence, the adjusted R
is used because it does not automatically increase as independent variables are added to
the model.
AIC and BIC
When evaluating a collection of models that explain the same dependent variable, we
cannot rely on the adjusted R2 alone. Two commonly used statistics for this purpose are
Akaike’s information criterion(AIC) and Schwarz’s Bayesian information criterion (BIC).
These stats are often provided as part of the regression software output.
Lower values of both measures are better.
When do we prefer one measure over the other?
 AIC is preferred if the model is used for prediction purposes
 BIC is preferred when the best goodness of fit is the goal
Testing joint hypotheis for coefficients

Hypothesis tests of a single coefficient
The hypothesis tests of a single coefficient in a multiple regression are identical to those in
a simple regression.
If we are testing simply whether a variable is significant in explaining the dependent
variable’s variation, the hypotheses are:
H0: bj = 0
Ha: bj ≠ 0
The statistical software produces the t-statistics and P-values for a test of the slope
coefficient against zero for each independent variable in the model. t-stats greater than the
critical value indicate that the variable is significant.
Joint F-test
The joint F-test is used to jointly test a subset of variables in a multiple regression, where
the “restricted” model is based on a narrower set of independent variables nested in the
broader “unrestricted” model. The null hypothesis is that the slope coefficients of all
independent variables outside the restricted model are zero.

It is calculated as:
(Sum of squares error restricted model − Sum of squares error unrestricted)/q
F=
Sum of squares error unrestricted model/(n − k − 1)
where: q is the number of restrictions, i.e. the number of variables omitted
If the calculated F-stat exceeds the critical F-value we can conclude that at least one of the
omitted variables is statically significant.
General linear F-test
The general linear F-test is an extension of the joint F-test, where we test the significance of
the whole regression equation. The null hypothesis is that the slope coefficients on all
independent variables are 0, against the alternative that at least one coefficient is different
from 0.
The F-stat is calculated as:
mean regression sum of squares MSR
F= =
mean squared error MSE
If the calculated F-stat exceeds the critical F-value we can conclude that at least one of the
slope coefficients is different from 0.
Forecasting using multiple regression

To forecast the value of a dependent variable using a multiple linear regression model,
follow these three steps:
1. Obtain estimates b ̂0 , b
̂1 , b
̂2 , … . , b
̂k of the regression parameters b0, b1, b2,…bk. The ̂
symbol indicates that the values are estimated.
2. Determine the assumed values of the independent variables X̂ ̂ ̂
1i , X 2i , … … , X ki
3. Compute the predicted value of the dependent variable, 𝑌𝑖 ̂ using the following equation:
̂0 + b
̂i = b
Y ̂1 X
̂1i + b̂2 X
̂ 2i + ⋯ … + b ̂k X
̂ ki
The level of uncertainty around the forecast of the dependent variable is called the
standard error of forecast. This forecast error depends on how well the independent
variables (X1, X2, X3…) were forecasted i.e. the sampling error, as well as the model error.
The larger the sampling error, the larger is the standard error of the forecast of Y and the
wider is the confidence interval.

LM03 Model Misspecification

Misspecified functional form
Whenever we estimate a regression, we must assume that the regression has the correct
functional form. This assumption can fail in several ways as shown in Exhibit 2.
Failures in Regression Explanation Consequence
Functional Form
Omitted variables One or more important variables May lead to
are omitted from the regression heteroskedasticity or serial
correlation
Inappropriate form of Ignoring a nonlinear relationship May lead to
variables between the dependent and heteroskedasticity
independent variable
Inappropriate variable One or more regression variables May lead to
scaling may need to be transformed heteroskedasticity or
before estimating the regression. multicollinearity
Inappropriate data Regression model pools data May lead to
pooling from different samples that heteroskedasticity or serial
should not be pooled correlation
Heteroskedasticity
There are two types of heteroskedasticity:
 Unconditional heteroskedasticity: When heteroskedasticity of the error variance is
not correlated with the independent variables. Unconditional heteroskedasticity is
not a problem.
 Conditional heteroskedasticity: The error variance is correlated with the values of
the independent variables. Conditional heteroskedasticity is a problem. It results in
underestimation of standard errors, so t-statistics are inflated and Type I errors are
more likely.
The figure below illustrates conditional heteroskedasticity.

Conditional heteroskedasticity can be detected using the Breusch–Pagan (BP) test. It can be
corrected by computing ‘robust standard errors’.
Serial correlation
In serial correlation, regression errors in one period are correlated with errors from
previous periods. It is often found in time-series regressions.
The following figure demonstrates the presence of serial correlation. When the previous
error term is positive the next error term is also most likely positive and vice versa.
Consequences:
Independent Variable Is Lagged Value Invalid Coefficient Invalid Standard Error
of Dependent Variable Estimates Estimates
No No Yes
Yes Yes Yes
Serial correlation can be detected using Breusch–Godfrey (BG) test. It can be corrected by

computing ‘robust standard errors’.
Multicollinearity
Multicollinearity may occur when two or more independent variables are highly correlated
or when there is an approximate linear relationship among independent variables.
Consequences: The standard errors are inflated so the t-stats of coefficients are artificially
small and cannot reject the null hypothesis.
Multicollinearity can be detected by using the variance inflation factor (VIF). VIF values
above 5 warrant further investigation. Whereas, VIF values above 10 indicate serious
multicollinearity problems.
It can be corrected by dropping one or more of the regression variables, using a different
proxy for one of the variables, or increasing the sample size.
Exhibit 14 summarizes the three issues arising from regression assumption violations.
Assumption Violation Issue Detection Correction
Homoskedastic Heteroskedastic Biased Visual Revise model;
error terms error terms estimates of inspection of use robust
coefficients’ residuals; standard errors
standard errors Breusch– Pagan
test
Independence Serial Inconsistent Breusch– Revise model;
of observations correlation estimates of Godfrey test use serial-
coefficients and correlation
biased standard consistent
errors standard errors
Independence Multicollinearity Inflated Variance Revise model;
of independent standard errors inflation factor increase sample
variables size

LM04 Extensions of Multiple Regression

Influence analysis
An influential observation is an observation whose inclusion may significantly alter
regression results. Two kinds of observations may potentially influence regression results:
 A high-leverage point: A data point having an extreme value of an independent
variable (X).
 An outlier: A data point having an extreme value of the dependent variable (Y).
Detecting Influential Points
Leverage (hii)
A high-leverage point can be identified using a measure called leverage (hii). Leverage
measures the distance between the value of the ith observation of an independent variable
and the mean value of that variable across all n observations. Leverage ranges from 0 to 1,
the higher the leverage the more distant the observation’s value is from the mean, and
hence the more influence it can exert on the estimated regression line. Statistical software
packages can easily calculate the leverage measure.
Interpretation:
If Then
𝑘+1 The observation is potentially influential
Leverage > 3 ( 𝑛
)
where: k is the number of independent variables

Studentized residual (ti*)
An outlier can be identified using a measure called studentized residuals. Statistical
software packages can calculate and present this measure.
Interpretation:
If Then
| ti*| > 3 Flag observation as being an outlier
| ti*| > critical value of the t-statistic with n-k-2 Flag outlier observation as being
degrees of freedom potentially influential
Cook’s distance
Cook’s distance, or Cook’s D (Di), is a metric for identifying influential data points. It
measures how much the estimated values of the regression change if observation i is
deleted from the sample.
Interpretation:
If Then
Di > 0.5 The ith observation may be influential and merits further investigation.
Di > 1.0 The ith observation is highly likely to be an influential data point.

Di > 2√𝑘/𝑛 The ith observation is highly likely to be an influential data point.
Dummy variables in multiple regression

Dummy (or indicator) variables represent qualitative independent variables. They take on
a value of 1 if a particular condition is true and 0 if that condition is false. To distinguish
among n categories, the model must include n – 1 dummy variables.
An intercept dummy adds to or reduces the original intercept if a specific condition is met.
When the intercept dummy is 1, the regression line shifts up or down parallel to the base
regression line.
Yi = b0 + d0Di + b1Xi + εi.
A slope dummy allows for the slope of the regression line to change if a specific condition is
met.
Yi = b0 + b1Xi + d1DiXi + εi.
It is also possible for a regression model to use both intercept and slope dummy variables.
Yi = b0 + d0Di + b1Xi + d1DiXi + εi.
Exhibit 11, Panel A, shows the effect of an intercept dummy variable.
Exhibit 11, Panel B, shows the effect of a slope dummy variable.

Exhibit 11, Panel C, shows the combined effect of a intercept and slope dummy variable.
Logistic regression model

Qualitative dependent variables are outcome variables describing data that fit into
categories (e.g. bankrupt or not bankrupt). A logistic transformation is applied when the
model contains qualitative dependent variables. The logistic transformation is:
ln[P/(1 − P)]
The logit transformation linearizes the relation between the transformed dependent
variable and the independent variables.

𝑃
𝑙𝑛 ( ) = 𝑏0 + 𝑏1 𝑋1 + 𝑏2 𝑋2 + 𝑏3 𝑋3 + €
1−𝑃
The natural logarithm (ln) of the odds of an event happening is the log odds which is also
called the logit function.
Logistic regression coefficients are typically estimated using the maximum likelihood
estimation (MLE) method rather than by least squares.
In a logit model, slope coefficients are interpreted as the change in the log odds that the
event happens per unit change in the independent variable, holding all other independent
variables constant.
A likelihood ratio (LR) test is a method to assess the fit of logistic regression models. The
test is similar to the joint F-test. It compares the fit of the restricted and unrestricted
models.

LM05 Time-Series Analysis

Time series
A time series is a set of observations on a variable measured over different time periods. A
time series model allows us to make predictions about the future values of a variable.
Linear vs log-linear trend models

 When the dependent variable changes at a constant amount with time, a linear trend
model is used.
The linear trend equation is given by yt = b0 + b1 t + εt , t = 1, 2, … , T
 When the dependent variable changes at a constant rate (grows exponentially), a log-
linear trend model is used.
The log-linear trend equation is given by ln yt = b0 + b1t, t = 1, 2, …, T
 A limitation of trend models is that by nature they tend to exhibit serial correlation in
errors, due to which they are not useful.
 The Durban-Watson statistic is used to test for serial correlation. If this statistic
differs significantly from 2, then we can conclude the presence of serial correlation in
errors. To overcome this problem, we use autoregressive time series (AR) models.
Requirements for a time series to be covariance stationary

AR models can only be used for time series that are covariance stationary.
A time-series is covariance stationary if it meets the following three conditions:
 The expected value of the time series (its mean) must be constant and finite in all
periods.
 The variance must be constant and finite in all periods.
 The covariance of time series, past or future must also be constant and finite in all
periods.
Autoregressive (AR) models

An autoregressive time series model is a linear model that predicts its current value using
its most recent past value as the independent variable. An AR model of order p, denoted by
AR(p) uses p lags of a time series to predict its current value.
xt = b0 + b1 x(t–1) + b2 x(t–2) + … + bp x(t– p) + εt
The chain rule of forecasting is used to predict successive forecasts.
The one-period ahead forecast of xt from an AR(1) model is x̂t+1 = b̂0 + b̂1 xt
xt+1 can be used to forecast the two-period ahead value : x̂t+2 = b̂0 + b̂1 xt+1

Autocorrelations of the residuals

 If an AR model has been correctly specified, the residual terms will not exhibit serial
correlation. We cannot use the Durban–Watson statistic to test for serial correlation in
AR models. Instead, we check the autocorrelations of the residuals.
 The autocorrelations of residuals are the correlations of the residuals with their own
past values. The autocorrelation between one residual and another one at lag k is
known as the kth order autocorrelation.
 If the model is correctly specified, the autocorrelation at all lags must be equal to 0.
 A t-test is used to test whether the error terms in a time series are serially correlated.
residual autocorrelation
Test stat =
standard error
Mean reversion
A time series is said to be mean-reverting if it tends to fall when its level is above its mean
and rise when its level is below its mean. If a time series is covariance stationary, then it
will be mean-reverting.
The mean-reverting level is calculated as:
b0
Mean– reverting level xt =
1 − b1
In- sample and out- of- sample forecasts, root mean squared error criterion (RMSE)
There are two types of forecasting errors based on the period used to predict values:
 In-sample forecast errors: these are residuals from a fitted series model used to
predict values within the sample period.
 Out-of-sample forecast errors: these are regression errors from the estimated
model used to predict values outside the sample period. Out-of-sample analysis is a
realistic way of testing the forecasting accuracy of a model and aids in the selection
of a model.
Root mean squared error (RMSE), square root of the average squared forecast error, is
used to compare the out-of-sample forecasting performance of the models. If two models
are being compared, the one with the lower RMSE for out-of-sample forecasts has better
forecast accuracy.
Instability of coefficients
Estimates of regression coefficients of the time-series model can change substantially
across different sample periods used for estimating the model. When selecting a time
period:
 Determine whether economics or environment have changed.
 Look at graphs of the data to see if the time series looks stationary.

Most economic and financial time series data are not stationary.
Random walk
A random walk is a time series in which the value of the series in one period is the value of
the series in the previous period plus an unpredictable random error.
The equation for a random walk without a drift is:
xt = xt−1 + εt
The equation for a random walk with a drift is:
xt = b0 + xt−1 + εt
They do not have a mean-reverting level and are therefore not covariance stationary. For
example, currency exchange rates.
Unit root
 For an AR (1) model to be covariance stationary, the absolute value of the lag coefficient
b1 must be less than 1. When the absolute value of b1 is 1, the time series is said to have
a unit root.
 All random walks have unit roots. If the time series has a unit root, then it will not be
covariance stationary.
 A random-walk time series can be transformed into one that is covariance stationary by
first differencing the time series. We define a new variable y as follows:
yt = xt − xt−1 = εt , where E(εt ) = 0, E(ε2t ) = σ2 , E(εt εs ) = 0 if t ≠ s
 We can then use and AR model on the first-differenced series.
Unit root test
We can detect the unit root problem by using the Dickey-Fuller test.
It is a unit root test based on a transformed version of the AR (1) model xt = b0 + b1 xt−1 +
εt
Subtracting xt-1 from both the sides, we get
xt − xt−1 = b0 + (b1 − 1) xt−1 + εt or
xt − xt−1 = b0 + g1 xt−1 + εt
where 𝑔1 = 𝑏 − 1
If b1 = 1, then g1 = 0. This means there is a unit root in the model.
Seasonality
If the error term of a time-series model shows significant serial correlation at seasonal lags,
the time-series has significant seasonality. This means there is significant data in the error
terms that is not being captured in the model.
Seasonality can be corrected by including a seasonal lag in the model. For instance, to

correct seasonality in the quarterly time series, modify the AR (1) model to include a
seasonal lag 4:
xt = b0 + b1 x(t–1) + b2 x(t−4) + εt
If the revised model shows no statistical significance of the lagged error terms, then the
model has been corrected for seasonality.
Autoregressive conditional heteroskedasticity (ARCH)

 If the variance of the error in a time series depends on the variance of the previous
errors than this condition is called autoregressive conditional heteroskedasticity
(ARCH).
 If ARCH exists, the standard errors for the regression parameters will not be correct.
We will have to use generalized least squares or other methods that correct for
heteroskedasticity.
 To test for first-order ARCH, we regress the squared residual on the squared residual
from the previous period. ε̂2t = a0 + a1 ε̂2t−1 + ut
If the coefficient in our model is statistically significant, the time- series model has
ARCH(1) errors.
 If a time-series model has significant ARCH, then we can predict the next period error
variance using the formula:
̂2t+1 = â0 + â1 ε̂2t
σ
Working with two time series

If a linear regression is used to model the relationship between two time series, a test such
as the Dickey-Fuller test should be performed to determine whether either time series has
a unit root.
 If neither of the time series has a unit root, then we can safely use linear regression.
 If one of the two time series has a unit root, then we should not use linear
regression.
 If both time series have a unit root and they are cointegrated (exposed to the same
macroeconomic variables), we may safely use linear regression.
 If both time series have a unit root but are not cointegrated, then we cannot not use
linear regression.
The Engle-Granger/Dicky-Fuller test is used to determine if a time series is cointegrated.
Selecting an appropriate time-series model

Section 16.1 from the curriculum provides a step-by-step guide on selecting an appropriate
time-series model.
1. Understand the investment problem you have, and make an initial choice of model. One

alternative is a regression model that predicts the future behavior of a variable based
on hypothesized causal relationships with other variables. Another is a time-series
model that attempts to predict the future behavior of a variable based on the past
behavior of the same variable.
2. If you have decided to use a time-series model, compile the time series and plot it to see
whether it looks covariance stationary. The plot might show important deviations from
covariance stationarity, including the following:
• a linear trend;
• an exponential trend;
• seasonality; or
• a significant shift in the time series during the sample period (for example, a change
in mean or variance).
3. If you find no significant seasonality or shift in the time series, then perhaps either a
linear trend or an exponential trend will be sufficient to model the time series. In that
case, take the following steps:
• Determine whether a linear or exponential trend seems most reasonable (usually by
plotting the series).
• Estimate the trend.
• Compute the residuals.
• Use the Durbin–Watson statistic to determine whether the residuals have significant
serial correlation. If you find no significant serial correlation in the residuals, then
the trend model is sufficient to capture the dynamics of the time series and you can
use that model for forecasting.
4. If you find significant serial correlation in the residuals from the trend model, use a
more complex model, such as an autoregressive model. First, however, reexamine
whether the time series is covariance stationary. The following is a list of violations of
stationarity, along with potential methods to adjust the time series to make it
covariance stationary:
• If the time series has a linear trend, first-difference the time series.
• If the time series has an exponential trend, take the natural log of the time series
and then first-difference it.
• If the time series shifts significantly during the sample period, estimate different
time-series models before and after the shift.
• If the time series has significant seasonality, include seasonal lags (discussed in Step
7).
5. After you have successfully transformed a raw time series into a covariance-stationary
time series, you can usually model the transformed series with a short autoregression.
To decide which autoregressive model to use, take the following steps:
• Estimate an AR(1) model.

• Test to see whether the residuals from this model have significant serial correlation.
• If you find no significant serial correlation in the residuals, you can use the AR(1)
model to forecast.
6. If you find significant serial correlation in the residuals, use an AR(2) model and test for
significant serial correlation of the residuals of the AR(2) model.
• If you find no significant serial correlation, use the AR(2) model.
• If you find significant serial correlation of the residuals, keep increasing the order of
the AR model until the residual serial correlation is no longer significant.
7. Your next move is to check for seasonality. You can use one of two approaches:
• Graph the data and check for regular seasonal patterns.
• Examine the data to see whether the seasonal autocorrelations of the residuals from
an AR model are significant (for example, the fourth autocorrelation for quarterly
data) and whether the autocorrelations before and after the seasonal
autocorrelations are significant. To correct for seasonality, add seasonal lags to your
AR model. For example, if you are using quarterly data, you might add the fourth lag
of a time series as an additional variable in an AR(1) or an AR(2) model.
8. Next, test whether the residuals have autoregressive conditional heteroskedasticity. To
test for ARCH(1), for example, do the following:
• Regress the squared residual from your time-series model on a lagged value of the
squared residual.
• Test whether the coefficient on the squared lagged residual differs significantly from
0.
• If the coefficient on the squared lagged residual does not differ significantly from 0,
the residuals do not display ARCH and you can rely on the standard errors from
your time-series estimates.
• If the coefficient on the squared lagged residual does differ significantly from 0, use
generalized least squares or other methods to correct for ARCH.
9. Finally, you may also want to perform tests of the model’s out-of-sample forecasting
performance to see how the model’s out-of-sample performance compares to its in-
sample performance.
Using these steps in sequence, you can be reasonably sure that your model is correctly
specified.

LM06 Machine Learning

Supervised machine learning, unsupervised machine learning, and deep learning
Supervised machine learning makes use of labeled training data. It can be divided into
two categories:
 Regression: The target variable is continuous.
 Classification: The target variable is categorical or ordinal.
Unsupervised machine learning does not make use of labelled training data. The ML
program has to discover structure within the data on its own. Two important types of
problems well suited to unsupervised ML are:
 Dimension reduction: Reducing the number of features (X variables).
 Clustering: Sorting observations into groups.
Deep learning refers to sophisticated algorithms which are used for highly complex tasks
such as image classification, face recognition, speech recognition, and natural language
processing.
Overfitting
Overfitting refers to an issue where the model fits training data perfectly but does not work
well with out-of-sample data.
There are two methods to reduce overfitting:
 Preventing the algorithm from getting too complex: This is based on the principle
that the simplest solution often tends to be the correct one.
 Cross-validation: This is based on the principle of avoiding sampling bias. A
commonly used technique is k-fold cross-validation. Here the data is shuffled
randomly and then divided into k equal sub-samples, with k-1 samples used as
training samples and one sample, the kth, used as a validation sample.
The total out-of-sample error can be decomposed into:
 Bias error: refers to the degree to which a model fits the training data. Underfitted
models have high bias errors.
 Variance error: refers to how much the model’s result change in response to new
data. Overfitted models have high variance errors.
 Base error: refers to errors due to randomness in the data.
An overfitted model will have a low bias error but a high variance error.
Generalization refers to the degree to which a model retains its explanatory power when
predicting out-of-sample. A model that generalizes well has low variance error.

Supervised machine learning algorithms

In penalized regression, the regression coefficients are chosen to minimize the sum of
squared residuals plus a penalty term that increases with the number of included variables.
So, a feature must make a sufficient contribution to model fit to offset the penalty from
including it. Because of this penalty, the model remains parsimonious and only the most
important variables for explaining Y remain in the model. A popular type of penalized
regression is LASSO.
Support vector machine (SVM) is a linear classifier that aims to seek the optimal
hyperplane – the one that separates the two sets of data points by the maximum margin.
K-nearest neighbor (KNN) classifies a new observation by finding similarities
(“nearness”) between it and its k-nearest neighbors in the existing data set.
Classification and regression tree (CART) can be applied to predict a categorical variable
or a continuous target variable. A binary CART tree is a combination of an initial root node,
decision nodes, and terminal nodes. The root node and each decision node represent a
single feature (f) and a cutoff value (c) for that feature. The CART algorithm iteratively
partitions the data into sub-groups until terminal nodes are formed that contain the
predicted label.
A random forest classifier is a collection of many decision trees generated by a bagging
method or by randomly reducing the number of features available during training.
In ensemble learning, we combine predictions from a collection of models. This method
typically produces more accurate and more stable predictions than the best single model.
Unsupervised machine learning algorithms

Principal components analysis (PCA) is used to reduce highly correlated features into a
few uncorrelated composite variables. A composite variable is a variable that combines two
or more variables that are statistically strongly related to each other.
K-means algorithm repeatedly partitions observations into k non-overlapping clusters.
The number of clusters, k, is a hyperparameter whose value must be set by the researcher
before learning begins. Each cluster is characterized by its centroid and each observation is
assigned to the cluster with the centroid to which that observation is closest.
Hierarchical clustering algorithms create intermediate rounds of clusters in increasing or
decreasing size until a final clustering is reached. Agglomerative clustering (or bottom-up)
hierarchical clustering begins with each observation being treated as its own cluster.
Divisive clustering (or top-down) hierarchical clustering starts with all the observations
belonging to a single cluster.

Neural networks, deep learning nets, and reinforcement learning

Neural networks have layers of nodes connected by links. The three types of layers are:
input layer, hidden layer and output layer. Learning takes place in the hidden layer through
improvements in the weights applied to nodes.
Neural networks with many hidden layers (at least 3 but often more than 20 hidden layers)
are known as deep learning nets (DLNs). NNs and DLNs have been successfully applied to
a wide variety of complex tasks characterized by non-linearities and interactions among
features, particularly pattern recognition problems.
Reinforcement learning (RL) algorithm involves an agent that should perform actions
that will maximize its reward over time, taking into consideration the constraints of its
environment.

LM07 Big Data Projects

Steps in a Data Analysis Project
Exhibit 1 from the curriculum shows the steps involved in executing a data analysis project.
The steps vary depending on whether we are working with structured data (data in tabular
format) or unstructured data (textual data).
Conceptualization: Define what the output of the model should be, (for example, will stock
prices go up/down in a week from now), how this model will be used, who will use this
model, and how this model will become part of the existing business process.
Text problem formulation: Define the text classification problem, identify the exact inputs
and outputs of the model. For example, computing sentiment scores (positive, negative,
neutral) from the text data.
Data Prep & Wrangling

Structured data
For structured data, data preparation and wrangling involve data cleansing and data
preprocessing.
Data cleansing involves resolving:
 Incompleteness errors: Data is missing.
 Invalidity errors: Data is outside a meaningful range.
 Inaccuracy errors: Data is not a measure of true value.
 Inconsistency errors: Data conflicts with the corresponding data points or reality.
 Non-uniformity errors: Data is not present in an identical format.

 Duplication errors: Duplicate observations are present.

Data preprocessing involves performing the following transformations:
 Extraction: A new variable is extracted from the current variable for ease of
analyzing and using for training the ML model.
 Aggregation: Two or more variables are consolidated into a single variable.
 Filtration: Data rows not required for the project are removed.
 Selection: Data columns not required for the project are removed.
 Conversion: The variables in the dataset are converted into appropriate types to
further process and analyze them correctly.
Scaling is the process of adjusting the range of a feature by shifting and changing the
scale of data. Two common methods used for scaling are:
Xi −Xmin
o Normalization: Xi(normalized) = X
max −Xmin
Xi −µ
o Standardization: X i(standarized) = σ
Unstructured data
For unstructured data, data preparation and wrangling involve a set of text-specific
cleansing and preprocessing tasks.
Text cleansing involves removing the following unnecessary elements from the raw text:
 Html tags
 Most punctuations
 Most numbers
 White spaces
Text preprocessing involves performing the following transformations:
 Tokenization: the process of splitting a given text into separate tokens where each
token is equivalent to a word.
 Normalization: The normalization process involves the following actions:
o Lowercasing – Removes differences among the same words due to upper and
lower cases.
o Removing stop words – Stop words are commonly used words such a ‘the’,
‘is’ and ‘a’.
o Stemming – Converting inflected forms of a word into its base word.
o Lemmatization – Converting inflected forms of a word into its morphological
root (known as lemma). Lemmatization is a more sophisticated approach as
compared to stemming and is difficult and expensive to perform.
 Creating bag-of-words (BOW): It is a collection of distinct set of tokens that does not
capture the position or sequence of the words in the text.
 Organizing the BOW into a Document term matrix (DTM): It is a table, where each
row of the matrix belongs to a document (or text file), and each column represents a
token (or term). The number of rows is equal to the number of documents in the

sample dataset. The number of columns is equal to the number of tokens in the final
BOW. The cells contain the counts of the number of times a token is present in each
document.
 N-grams and N-grams BOW: In some cases, a sequence of words may convey more
meaning than individual words. N-grams is a representation of word sequences. The
length of a sequence varies from 1 to n. A one-word sequence is a unigram; a two-
word sequence is a bigram, and a 3-word sequence is a trigram; and so on.
Data Exploration
Exploratory data analysis (EDA) is a preliminary step in data exploration that involves
summarizing and observing data. The objectives of EDA include:
 Serving as a communication medium among project stakeholders
 Understanding data properties
 Finding patterns and relationships in data
 Inspecting basic questions and hypotheses
 Documenting data distributions and other characteristics
 Planning modeling strategies for the next steps
Visualization techniques for EDA include: histograms, box plots, scatterplots, word clouds,
etc.
In general, EDA helps identify the general trends in the data as well as the relationships
between data. These relationships and trends can be used for feature selection and feature
engineering.
Feature selection is the process of selecting only the relevant features from the data set to
reduce model complexity. Feature selection methods used for text data include:
 Term frequency (TF): Features with very high or very low term frequencies are
removed. These represent noisy features.
 Document frequency (DF): The DF of a token is defined as the number of documents
(texts) that contain the respective token divided by the total number of documents.
This measure also helps in identifying and removing noisy features.
 Chi-square test: This test allows us to rank tokens by their usefulness to each class
in text classification problems. Features with high scores can be retained whereas
features with low scores can be removed.
 Mutual information measure: This measure tells us how much information is
contributed by a token to a class of texts. The MI score ranges from 0 – 1. A score
close to 0 indicates that the token is not very useful and can be removed. Whereas, a
score close to 1 indicates that the token is closely associated with a particular class
and should be retained.
Feature engineering is the process of creating new features by changing or transforming
existing features. Feature engineering for text data includes:

 Converting numbers into tokens: To preserve the meaning conveyed by numbers of

different length they are converted into different tokens. For example, four-digit
numbers are replaced with “/number4/,” 10-digit numbers with “/number10/,”.
 Creating n-grams: Sometimes multi-word sequences can carry more meaning than
an individual word. N-grams are used to keep the connection between words intact.
 Using name entity recognition (NER): NER is an algorithm that allows us to identify
the names of organizations, date, money, time, etc. and tag certain tokens with these
names.
 Using parts of speech (POS): Similar to NER, POS algorithms allows us to tag each
token with the corresponding part of speech such as noun, verb, adjective, proper
noun etc.
Model Training
Model training consists of three major tasks: method selection, performance evaluation,
and model tuning.
Method selection: This decision is based on the following factors:
 Whether the data project involves labeled data (supervised learning)or unlabeled
data (unsupervised learning)
 Type of data: numerical, continuous, or categorical; text data; image data; speech
data; etc.
 Size of the dataset
Performance evaluation: Commonly used techniques are:
 Error analysis using confusion matrix: A confusion matrix is created with four
categories - true positives, false positives, true negatives and false negatives.
The following metrics are used to evaluate a confusion matrix:
Precision (P) = TP/(TP + FP)
Recall (R) = TP/(TP + FN)
Accuracy = (TP + TN)/(TP + FP + TN + FN)
F1 score = (2 * P * R)/(P + R)
The higher the accuracy and the F1 score, the better the model performance.
 Receiver Operating Characteristic (ROC): ROC curves and area under the curve
(AUC) of various models are calculated and compared. An area under the curve
(AUC) close to 1 indicates a perfect model. Whereas, an AUC of 0.5 indicated
random guessing. In other words, a more convex curve indicates better model
performance.
 Calculating root mean squared error (RMSE): The root mean squared error is
computed by finding the square root of the mean of the squared differences
between the actual values and the model’s predicted values (error). The model with
the smallest RMSE is the most accurate model.

Model tuning
 Involves managing the trade-off between model bias error (which is associated with
underfitting) and model variance error (which is associated with overfitting).
 ‘Grid search’ is a method of systematically training an ML model by using different
hyperparameter values to determine which values lead to best model performance.
 A fitting curve of in-sample error and out-of-sample error on the y-axis versus
model complexity on the x-axis is useful for managing the trade-off between bias
and variance errors.
Evaluating the Fit of a ML algorithm

Fitting describes the degree to which an ML model can be generalized to new data.
 Underfit: Model does not fit the training data well.
 Overfit: Model is complex and fits the training data too well. It is unlikely to perform
well on out-of-sample data.
 Good fit: Model fits the training data well and yet is simple enough so that it can
generalize well to out-of-sample data.

Financial Statement Analysis 2023 Level II High Yield Notes
LM01 Intercorporate Investments

Basic corporate investment categories
The table below summarizes the accounting treatment for various investment categories
under IFRS.
Financial Assets Associates Business Joint
Combinations Ventures
Influence Not significant Significant Controlling Shared
control
Typical Usually < 20% Usually 20% Usually > 50% or
percentage to 50% other indications of
interest control
Financial Classified as: Equity Consolidation IFRS:
Reporting  Fair value through profit method Equity
or loss method
 Fair value through other
comprehensive income
 Amortized cost
Investments in financial assets: IFRS 9
The following flowchart will help understand the classification of financial assets under
IFRS 9.

Reclassification under IFRS 9:

 The reclassification of equity instruments is not permitted because the initial
classification of FVPL and FVOCI is irrevocable.
 Reclassifications of debt instruments from FVPL to amortized costs (or vice versa)
are permitted only when the business model changes in a way that significantly
affects operations. Changes to the business models require judgment and are
expected to be infrequent.
Investments in associates and joint ventures
An investment is considered an “associate company” when the investor has (or can
exercise) significant influence, but not control, over the investee’s business activities.
Significant influence is presumed with 20 – 50% ownership or voting power of the
associate (investee). Significant influence may be evidenced by:
 Representation on the board of directors
 Participation in the policy-making process
 Material transactions between the investor and the investee
 Interchange of managerial personnel
 Technological dependency
Joint ventures are ventures undertaken and controlled by two or more parties. IFRS
identifies two characteristics of joint ventures as:
 A contractual arrangement exists between two or more venturers.

 The contractual arrangement establishes joint control.

The equity method is used to account for investments in associates and joint ventures.
Equity method: a one-line consolidation method.
Balance sheet:
 Investment account reflected as a single line item (non-current asset) on the balance
sheet.
 Equity investment is initially recorded on the investor’s balance sheet at cost. It has
three components:
o Reported value on investee B/S
o Fair value surplus
o Goodwill
 Subsequently, the carrying amount of the investment is adjusted to recognize the
investor’s proportionate share of the investee’s earnings or losses.
 Value of investment = beginning value + share of profits – share of dividends
 Dividends received from the investee are treated as a return of capital and reduce
the carrying amount of the investment. They are not reported on the income
statement.
 If the investment costs exceed the book value of the investee, then we amortize the
excess purchase price attributable to plant and equipment.
 Value of investment = beginning value + share of profits – share of dividends -
amortization of excess purchase price attributable to plant and equipment
Income statement:
 The share of income (not dividends) is recorded in the investor’s income statement
Transactions with the investee:
 Because an investor company can influence the terms and timings of transactions
with associates, profits from such transactions cannot be realized until confirmed
through use of a third-party sale.
 Investor company’s share of any unrealized profit must be deferred by reducing the
amount recorded under the equity method.
 This deferred profit is added back to the equity income when confirmed.
Business combinations
Under IFRS, there is no distinction between business combinations. Under US GAAP, there
are four types: merger, acquisition, consolidation, and variable interest entity.
The acquisition method is used for business combinations.
Acquisition method

The major points related to the acquisition method are listed below:
 Identifiable tangible and intangible assets (brand names, patents, etc.) and liabilities
of the acquired company are measured at fair value on the date of the acquisition.
 Assets and liabilities that were not previously recognized by the acquiree must be
recognized by the acquirer.
 The acquirer must recognize any contingent liability if it can be reliably measured
and the obligation arises from past events.
 The acquirer must recognize an indemnification asset on the acquisition date.
 Recognition of goodwill differs between IFRS and US GAAP:
o IFRS has two options for goodwill: partial goodwill and full goodwill.
Partial goodwill = fair value of the acquisition - acquirer’s share of the fair
value of all acquiree’s assets and liabilities
Full goodwill = fair value of the entity as a whole - the fair value of all
acquiree’s assets and liabilities
o US GAAP allows only full goodwill.
 In an acquisition, when the purchase price is less than the fair value of the target’s
(acquiree’s) net assets, the acquisition is considered to be a bargain acquisition.
IFRS and US GAAP require the difference between the fair value of the acquired net
assets and the purchase price to be recognized immediately as a gain.
Comparison of methods
The following table compares the effect of equity method and acquisition method on
financial ratios.
Equity Method Acquisition Method Acquisition Method
(Partial Goodwill) (Full Goodwill)
Profit Margin More favorable Less favorable than equity method
Leverage Most favorable Less favorable than More favorable than
equity method partial goodwill method
ROE Most favorable Less favorable than Less favorable than

ROA Most favorable Less favorable than Less favorable than

Current Ratio More favorable Less favorable
Asset Turnover Need to calculate


LM02 Employee Compensation: Post Employment and Share-Based

Types of post-employment benefit plans
Defined contribution plans:

 Defined contribution pension plans only specify the amount of periodic contribution
made to a plan.
 The final amount of the pension benefit to the employee will depend on how well
the investments perform.
 The periodic contributions are shown as an expense on the income statement.
 If there are any unpaid contributions they are shown as a liability on the balance
sheet.
Defined benefit plans:
 Defined benefit pension plans specify the amount of the pension benefit. The final
amount of the benefit to the employee depends on factors such as length of service
and final salary.
 Both IFRS and US GAAP require a pension plan’s funded status to be reported on the
balance sheet.
Funded status = PV of the defined benefit obligation - Fair value of the plan assets
 If the plan has a deficit, the underfunded amount is reported as a net pension
liability.
 If the plan has a surplus, the overfunded amount is reported as a net pension asset.
(Note: The amount of reported assets is subject to a ceiling. This is defined as the
present value of future economic benefits.)
Other post-employment benefits plans:
 Other post-employment benefits (OPB) are promises by a company to pay benefits
in the future, such as life insurance premiums or health care insurance for its
retirees.
 OPB are typically classified as DB plans and the accounting treatment is similar to
DB plans.
 In an OPB plan, the employer bears investment risk.
 Companies typically do not pre-fund OPB obligations. However, regulations vary
from country to country.
Components of pension costs
Total periodic pension costs (TPPC) is equal to the contributions plus change in the
pension liability during the year.
Total periodic pension costs = net pension liability at the end of the period – net pension

liability at the start of the period + employer contribution

Each period, the periodic pension cost is recognized in profit or loss (P&L) and/or in other
comprehensive income (OCI).
Under IFRS, the periodic pension cost is viewed as having three components:
IFRS Component IFRS Recognition
Service costs Recognized in P&L.
Recognized in P&L as the following amount: Net
Net interest income/expense
pension liability or asset × interest ratea
Recognized in OCI and not subsequently
amortized to P&L
Net return on plan assets = Actual return – (Plan
Remeasurements: Net return on plan
assets × Interest rate)
assets and actuarial gains and losses
Actuarial gains and losses = Changes in a
company’s pension obligation arising from
changes in actuarial assumptions
Under US GAAP, the periodic pension is viewed as having five components.
US GAAP Component US GAAP Recognition
Current service costs Recognized in P&L
Recognized in OCI and subsequently amortized to
Past service costs
P&L over the service life of employees
Interest expense on pension Recognized in P&L
obligation
Recognized in P&L as the following amount: Plan
Expected return on plan assets
assets × expected return
Recognized immediately in P&L or, more
commonly, recognized in OCI and subsequently
amortized to P&L using the corridor or faster
recognition method.b
Actuarial gains and losses including
Difference between expected and actual return
differences between the actual and
on assets = Actual return – (Plan assets ×
expected returns on plan assets
Expected return)
Actuarial gains and losses = Changes in a
company’s pension obligation arising from
changes in actuarial assumptions
a The interest rate used here is equal to the discount rate used to measure the pension liability (the yield on
high-quality corporate bonds).
b If the cumulative amount of unrecognized actuarial gains and losses exceeds 10 percent of the greater of the
value of the plan assets or of the present value of the DB obligation (under US GAAP, the projected benefit

obligation), the difference must be amortized over the service lives of the employees.
Current service cost: amount by which pension obligation increases as a result of

employees’ service in the current period.
Past service cost: amount by which pension obligation relating to employees’ service in
prior periods changes as a result of plan amendments or a plan curtailment.
Example:
Assume that the following information is provided about a company reporting under IFRS:
 PVDBO at start of year = 140
 FVPA at start of year = 120
 Discount rate = 10%
 Actual return on assets = 15
 Service costs = 7
 Increase in pension obligation due to changes in actuarial assumptions = 4
The components of pension costs can be calculated as:
Service cost = 7 (given)
Net interest income/expense = net pension liability or asset × interest rate = (140 - 120) x
10% = 2
Remeasurements:
Net return on plan assets = actual return – (plan assets × interest rate) = 15 – (120 x 10%)
=3
Actuarial gains and losses = 4 (given)
Remeasurements = - 3 + 4 = 1 (3 is subtracted as it will reduce the pension obligation)
Total periodic pension cost = 7 + 2 + 1 = 10
Impact of key DB pension assumptions
Impact of Assumption on Impact of Assumption

Assumption
Balance Sheet on Periodic Cost
Periodic pension costs will
typically be lower because
Higher discount rate Lower obligation
of lower opening obligation
and lower service costs
Higher rate of
Higher obligation Higher service costs.
compensation increase

No effect, because fair value of Not applicable for IFRS

Higher expected return
plan assets is used on balance Lower periodic pension
on plan assets
sheet expense under US GAAP
Classification of periodic pension costs
Pension costs are shown as operating expenses; however, conceptually, the interest
component and asset return component are non-operating.
Thus, to better reflect a company’s operating performance we can make the following
adjustments: show the service cost as an operating expense, add the interest component to
the company’s interest expense, and show the return on plan assets as non-operating
income.
The following example illustrates these adjustments.
Example:
US GAAP Income statement:
Rev = 100
Op Ex = (40)
EBIT = 60
Int Exp = (10)
Int Inc. = 9
PBT = 59
Pension disclosure:
Current service cost = (5)
Interest cost = (2)
Expected return on plan assets = 6
Actual return on plan assets = 8
Adjustments:
Total periodic pension cost = (5) + (2) + 6 = (1)
Op Ex = (40) – (1) + (5) = (44)
Int Exp = (10) + (2) = (12)
Int Inc. = 9 + 8 = 17
Adjusted PBT = 100 + (44) + (12) + 17 = 61.
Note: The difference between the adjusted PBT and the original PBT = 61 – 59 = 2,
corresponds with the difference between the actual return on plan assets and the expected
return on plan assets = 8 – 6 = 2

Cash flow information
If a firm’s contribution > economic pension expense, the contribution in excess of the
economic pension expense is a form of repayment.
Similarly, if a firm’s contribution < economic pension expense, the deficit (economic
pension expense – firm’s contribution) is a form of borrowing.
Example: You are given the following data:
Total pension cost = 100
Company’s contribution = 120
Tax rate = 25%
What is the after-tax amount by which company’s contribution exceeds pension cost?
Ans: (120 – 100) x (1 – 0.25) = 15
Before adjustment, CFO = inflow of 300 and CFF = outflow of 200
What is the CFO and CFF after adjustment?
Ans: Adjusted CFO inflow = 315; and adjusted CFF outflow = 215
Share-based compensation
Share-based compensation is intended to align incentives of management and owners.

From an accounting perspective, share-based compensation is treated as an expense even
though no cash changes hands when stock options/grants are issued.
Both IFRS and US GAAP require us to come up with a fair value to measure the share-based
compensation. In addition, the nature and extent of share-based compensation and its
effect on financial statements need to be disclosed.
There are several disadvantages of share-based compensation:
 Employees have limited influence over the company’s market value. Therefore, it
does not necessarily provide the desired alignment of incentives.
 Increased ownership may lead managers to be more risk averse or less risk averse
than they should be.
 When shares are granted to employees, the existing shareholder’s ownership is
diluted.
Stock grants: There are three types of stock grants:
 Outright grants
 Restricted grants: shares have to be returned if certain conditions are not met.
 Performance shares: determined by performance other than a change in share price.
To account for stock grants:
 Compensation expense is reported on the basis of fair value of the stock on the grant

date.
 Compensation expense is allocated over the service period.
Stock options
Compensation expense related to stock options is reported at fair value. The fair value has
to be estimated using an appropriate valuation model.
The fair value of stock options is sensitive to inputs and assumptions of the valuation
model:
 Exercise price up  value of option down
 Stock price volatility up  value of option up
 Estimated life of each award up  value of option up
 Estimated dividend yield up  value of call option down
 Risk free rate up  value of call option up
In accounting for stock options, there are several important dates, including the grant date,
the vesting date, the exercise date, and the expiration date.
 The grant date is the day that options are granted to employees.
 The vesting date is the date that employees can first exercise the stock options.
 The exercise date is the date when employees actually exercise the options and
convert them to stock.
 If the options go unexercised, they may expire at some pre-determined future date,
commonly 5 or 10 years from the grant date.
The compensation expense is allocated over the service period.
Example:
A company awards 1,000,000 stock options to its executives on 1 July 2015. The estimated
cost of each option is $0.50. The options require a service period of 4 years after the grant
date before vesting. What is the stock option expense for 2015?
Solution:
Total expense = 1,000,000 x 0.5 = $500,000
Expense per year = 500,000 / 4 = $125,000
Expense for 2015, from 1 July 2015 to 31 Dec 2015 i.e half a year = 125,000 / 2 = $62,500

LM03 Multinational Operations

Presentation currency, functional currency and local currency
Presentation (reporting) currency: The currency in which financial statement amounts

are presented. It is generally the currency of the parent company’s residence.
Local currency: The national currency of the country where the entity is located.
Functional currency: The currency of the primary economic environment in which an
entity operates. The factors that need to be considered while determining the functional
currency of subsidiary are:
1. The currency that mainly influences sales prices for goods and services.
2. The currency of the country whose competitive forces and regulations mainly
determine the sales price of its goods and services.
3. The currency that mainly influences labor, material, and other costs of providing
goods and services.
4. The currency in which funds from financing activities are generated.
5. The currency in which receipts from operating activities are usually retained.
Foreign currency transactions
Transaction exposure related to imports and exports can be summarized as follows:

Import purchase: Here the importer is required to pay in the foreign currency and the
settlement date falls after the purchase date. The importer is exposed to FX risk as the
functional currency may weaken against the foreign currency. This will increase the
amount of the functional currency that the importer is required to pay.
Export sales: Here the exporter agrees to be paid in foreign currency and the settlement
date falls sometime after the sale date. The exporter is exposed to FX risk as the functional
currency may strengthen against the foreign currency. This will decrease the amount of
functional currency that the exporter will receive.
Both U.S. GAAP and IFRS require the change in the value of foreign currency asset or
liability resulting from foreign currency transaction to be treated as a gain or loss reported
on the income statement.
Translation methods
Many companies have operations in foreign countries. Subsidiaries located in foreign

countries usually prepare their financial statements in the local currency. IFRS and US
GAAP require parent companies to prepare consolidated financial statements in which the
assets, liabilities, revenues, and expenses of both domestic and foreign subsidiaries are
added to those of the parent company. To prepare consolidated financial statements,

parent companies must translate the foreign currency financial statements into the
parent’s presentation currency.
If the foreign subsidiary’s functional currency is different from the parent’s presentation
currency, the current rate method must be used.
If the foreign subsidiary’s functional currency is same as the parent’s presentation
currency, the temporal method must be used.
Current rate method:
Under this method, the foreign entity’s foreign currency financial statements are translated
into the parent’s presentation currency using the following procedures:
1. All assets and liabilities are translated at the current exchange rate at the balance
sheet date.
2. Stockholders’ equity accounts are translated at historical exchange rates.
3. Revenues and expenses are translated at the exchange rate that existed when the
transactions took place (For practical reasons, a rate that approximates the
exchange rates at the dates of the transactions, such as an average exchange rate,
may be used).
4. Net translation gain/loss is unrealized (will be realized when the entity is sold) and
is not shown on the income statement. Translation adjustment is reflected in equity.
Temporal method:
Under this method, the foreign entity’s foreign currency financial statements are translated
into the parent’s presentation currency using the following procedures:
1. Assets and Liabilities
a. Monetary assets and liabilities are translated at the current exchange rate.
b. Non-monetary assets and liabilities measured at historical cost are
translated at historical exchange rates.
c. Non-monetary assets and liabilities measured at current value are translated at
the exchange rate at the date when the current value was determined.
2. Stockholders’ equity accounts are translated at historical exchange rates.
3. Revenue and Expenses
a. Revenues and expenses, other than those expenses related to non-monetary
assets, are translated at the exchange rate that existed when the transactions
took place.
b. Expenses related to non-monetary assets, such as cost of goods sold
(inventory), depreciation (fixed assets), and amortization (intangible
assets), are translated at the exchange rates used to translate the related
assets.
4. The translation gain/loss is shown on the income statement.

Instructor’s note: Items highlighted in bold indicate the differences between the temporal
method and the current rate method
Comparison of current rate and temporal methods
Current Rate Method Temporal Method
Monetary assets and liabilities Current rate Current rate
Non-monetary assets and liabilities Current rate Current rate

measured at current value
Non-monetary assets and liabilities Current rate Historical rates
measured at historical cost
Equity (other than retrained earnings) Historical rates Historical rates
Equity (retained earnings) Beginning balance plus translated net income

less dividends translated at historical rate
Revenue Average rate Average rate
Most expenses Average rate Average rate
Expenses related to assets translated Average rate Historical rates

at historical exchange rate
Treatment of the translation Accumulated as a Included as gain or loss
adjustment in the parent’s separate component of in net income
consolidated financial statements equity
Effect of currency exchange rate movement on financial statements
If value of assets to be translated is greater than value of liabilities to be translated, then we

have a net asset exposure. With the current rate there will almost always be a net asset
exposure.
The following table summarizes the effect of currency exchange rate movements on
financial statements:
Temporal Method, Net
Temporal Method Net
Monetary Liability Current Rate Method
Monetary Asset Exposure
Exposure

Foreign currency ↑Revenues ↑Revenues ↑Revenues

strengthens relative to ↑Assets ↑Assets ↑Assets
parent’s presentation ↑Liabilities ↑Liabilities ↑Liabilities
currency ↓Net income ↑Net income ↑Net income
↓Shareholder’s equity ↑Shareholder’s equity ↑Shareholder’s equity
Translation loss Translation gain Positive translation
adjustment
Foreign currency ↓Revenues ↓Revenues ↓Revenues
weakens relative to ↓Assets ↓Assets ↓Assets
parent’s presentation ↓Liabilities ↓Liabilities ↓Liabilities
currency ↑Net income ↓Net income ↓Net income
↑Shareholder’s equity ↓Shareholder’s equity ↓Shareholder’s equity
Translation gain Translation loss Negative translation
adjustment
Translation in a hyperinflationary economy
IFRS and US GAAP differ substantially in their approach to translating the foreign currency
financial statements of a subsidiary in a hyperinflationary economy.
US GAAP requires that the financial statements be translated using the temporal
method.
IFRS require financial statements to be first restated for inflation and then inflation
adjusted financial statements be translated at the current exchange rate.
The following procedures should be followed while adjusting the financial statements for
inflation:
 Monetary assets and liabilities are not restated.
 Nonmonetary assets and liabilities are adjusted for inflation.
o Restated costs are determined by applying the change in general price index
to the historical costs.
o If items are carried at revalued amounts (e.g. PP&E), these are restated from
the date of revaluation.
 All components of stockholders’ equity are restated by applying the change in
general price index from the beginning of the period or from the date of
contribution to the balance sheet.
 All income statement items are restated by applying the change in the general price
index from the dates when the items were originally recorded.
 Purchasing power gains/losses that arises from holding monetary assets and
monetary liabilities are included in net income. (Net monetary assets result in a
purchasing power loss; net monetary liabilities result in a purchasing power gain.)

Multinational operations and effective tax rate
Generally, multinational companies have to pay income taxes in the country in which the
profit is earned. Transfer prices (prices charged on intercompany transactions) affect the
allocation of profits between companies.
An entity with operations in multiple countries with different tax rates could aim to set
transfer prices such that a higher portion of its profit is allocated to lower tax rate
jurisdictions. To prevent this, countries have established various laws and practices to
prevent aggressive transfer pricing practices.
Whether and when a company also pays income taxes in its home country depends on the
specific tax regime.
An analyst can obtain information about the tax impact of multinational operations from
companies’ disclosure on effective tax rates.
Sustainability of earnings
For a multinational company, sales growth is driven not only by changes in volume and
price, but also by changes in the exchange rates. Growth in sales that comes from changes
in volume or price is more sustainable than growth in sales that comes from changes in
exchange rates.

LM04 Analysis of Financial Institutions

What makes financial institutions different from other companies?
The different types of financial institutions are:

 Banking industry – commercial banks, credit unions, etc.
 Insurance industry – property & casualty, life & health, reinsurance
 Investment management industry – mutual funds, exchange traded funds, hedge
funds, brokers & dealers.
Financial institutions have a high level of systemic risk. Their smooth functioning is crucial
to the overall economy. Due to the high level of systemic risk, the level of regulation in the
financial sector tends to be high.
The liabilities of financial institutions especially banks are made up primarily of deposits
and assets are predominantly financial assets. These assets are exposed to credit, market,
and liquidity risks.
Systematic risk: risk of disruption to financial services that is 1) caused by an impairment
of all or parts of the financial system and 2) has the potential to have serious negative
consequences for the economy as a whole.
Financial contagion: situation in which financial shocks spread from their place or sector
of origin to other locales or sectors.
Financial regulations of financial institutions
The objectives of global and regional regulatory bodies are:

 Minimize systematic risk
 Harmonize and globalize regulatory rules, standards, and oversight
 Ensure consistency of standards and regulations
 Minimize regulatory arbitrage
Some important financial regulators are: The Financial Stability Board, The International
Association of Deposit Insurers, The International Association of Insurance Supervisors,
The International Organization of Securities Commissions, Basel Committee on Banking
Supervision.
One of the most important organization focused on financial stability is the ‘Basel
Committee on Banking Supervision’. Its members include central banks from all over the
world, as well as several regulators. The key output of this committee is the international
regulatory framework for banks called – ‘Basel III’.
The objectives of this framework are to:
 Improve the banking sector’s ability to absorb shocks arising from financial and

economic stress.
 Improve risk management and governance of banks.
 Strengthen banks’ transparency and disclosures.
The three major highlights of the Basel III framework are:
 Minimum capital requirement: A bank must have sufficient equity capital to absorb
the loss in value of assets in a financial crisis.
 Minimum liquidity: A bank must hold enough high-quality liquid assets to cover its
liquidity needs in a 30-day liquidity stress scenario.
 Stable funding: A bank should have an adequate amount of stable funding relative to
the bank’s liquidity needs over a one- year horizon.
The CAMELS approach to analyzing a bank
The ‘CAMELS’ approach has six components which addresses the following questions:
1. Capital adequacy – Does the bank have sufficient capital given its assets?
2. Asset quality – What is the quality of the bank’s financial assets?
3. Management capabilities – Is the management effective? What is its track record?
4. Earnings – What is the level of earnings? What is the quality of earnings? Are
earnings trending up or down?
5. Liquidity: How strong is the liquidity position of the bank? What are the sources of
funding? Are the sources of funding stable?
6. Sensitivity to market risk: How sensitive are the bank’s earnings to market risk?
For each component a rating is assigned on a scale of 1 to 5 (where 1 is the best rating and
5 is the worst). After the components are rated, weights are assigned, and a weighted
average is taken to calculate the overall CAMELS score.
Capital adequacy
Capital adequacy is based on the portion of assets funded by capital. It indicates whether a
bank has enough capital to absorb losses without severely damaging its financial position.
Basel III’s minimum capital requirements are:
 Common Equity Tier 1 Capital must be at least 4.5% of risk weighted assets
 Total Tier 1 Capital must be at least 6.0% of risk-weighted assets
 Total Capital must be at least 8.0% of risk-weighted assets
Common Equity Tier 1 Capital: This is the safest type of capital. It includes common stock,
issuance surplus related to common stock, retained earnings, OCI, and certain adjustments.
Other Tier 1 Capital: This includes instruments issued by bank which meet criteria such as:
 Subordinate to deposits and other debt obligations.
 No fixed maturity.

 Interest or dividend payments, if any, should be at bank’s discretion.

 Collectively, the above two types are called ‘Total Tier 1 Capital.’
Tier 2 Capital: This is capital which meets criteria such as:
 Subordinate to deposits and to general creditors of bank.
 Minimum maturity of 5 years.
 Collectively, the three types are called ‘Total Tier 1 and Tier 2 Capital’
Risk weighted assets: Assets are weighted based on risk. For example, cash has a weight of
0%, corporate loans may be given a weight of 100% and non-performing loans may be
given a weight of 150%.
Asset quality
Asset quality is based on credit quality and diversification. The credit policies and overall
risk management processes of a bank impact its asset quality.
The three major asset categories for a bank are:
1. Loans – Generally the The quality of loans depends on:
largest component of overall  Creditworthiness of borrowers – High
assets creditworthiness implies high asset quality.
 Adequacy of adjustments for expected loan losses
– If the adjustments are not sufficient, then it
implies low asset quality.
2. Investments in securities They are measured differently depending on
issued by other entities – for categorization. The investments can be measured at:
example, a bond issued by  Amortized cost
another company.  Fair value, where unrealized gains/losses are
recorded in OCI (FVOCI)
 Fair value, where unrealized gains/losses flow
through the income statement (FVTPL)
3. Highly liquid financial They are considered the best quality assets.
instruments – for example,
T-Bills
Asset quality can be evaluated based on:
• Composition of assets
• Credit quality
Some ratios used in the analysis are:
• <Asset type>/total assets
• Allowance for loan losses / non-performing loans
• Provision for loan losses / net loan charge-offs

Management capabilities
Management capability refers to bank management’s ability to identify and exploit profit
opportunities while managing risk.
Other factors related to management capabilities are: governance structure, compliance
with laws and regulations, internal controls, transparency of management
communications, and financial reporting quality.
Earnings
Earnings refers to a bank’s return on invested capital relative to cost of capital.
The major earnings components for banks are:
• Net interest income (the difference between the interest earned on loans and the
interest paid to its depositors)
• Service income
• Trading income
Of these sources, net interest income and service income are more stable, whereas trading
income is the most volatile.
While evaluating earnings we should consider:
• Composition of earnings
• Earnings quality
• Earnings trend
• Loan impairment allowances
• Fair value estimates
Liquidity
A bank’s failure to honor a current liability could have a systemic impact. Therefore, it is
extremely important for banks to maintain a strong liquidity position.
There are two major considerations when evaluating liquidity:
1. Liquid assets relative to near-term expected cash flows. This is measured by the LCR
ratio.
Liquidity coverage ratio (LCR) = highly liquid assets / expected cash outflows
2. Stability of the banks funding sources. This is measured by the NSFR ratio.
Net stable funding ratio (NSFR) = available stable funding / required stable funding
Basel III sets a target minimum of 100% for both ratios.
Some other factors to consider when evaluating liquidity position are:
 Concentration of funding
 Contractual maturity mismatch

Sensitivity to market risk

Sensitivity to market risk deals with how adverse market movements impact a bank’s
earnings and financial position; measured using trading portfolio VaR and credit portfolio
VaR.
Market risk is the exposure to changes in interest rates, exchange rates, equity prices, or
commodity prices.
All entities have exposure to market risk. However, for banks the exposure to changes in
market prices arise because of mismatches in the maturity of banks’ deposits and loans.
Mismatches could be with respect to:
 Maturity
 Repricing frequency
 Reference rates
 Currency
Other factors relevant to analysis of a bank
Banking-specific analytical considerations not addressed by CAMELS include:

 Government support
 Government ownership
 Mission of banking entity
 Corporate culture
Analytical considerations not addressed by CAMELS that are also relevant for any
company:
 Competitive environment
 Off-balance-sheet items
 Segment information
 Currency exposure
Analyzing a property & casualty insurance company
Operations: Products and Distribution

Property insurance protects against loss or damage to property. Casualty insurance also
called liability insurance protects against legal liability related to a third party.
The premiums are collected at the start of the insurance contract. These premiums are then
invested during the float period. The float period represents the time difference between
receiving premiums and making insurance payouts.
In order to minimize payouts, insurance companies follow a prudent underwriting process.
They charge a sufficient price for the risks borne by them. They also try to diversify their

risks by not concentrating on one kind of policy, market, or customer type.

Property and insurance policies are typically annual and the final cost to the company is
usually known within a year of the covered event. The policies are then renewed at the end
of the year.
The claims received by the P&C companies are unpredictable and lumpier.
Earnings characteristics
Macro view: The insurance industry is highly competitive, price-sensitive and cyclical.
Micro view: The underwriting cycle is driven by expenses. When expenses are high relative
to premiums, firms exit the market. On the other hand, when expenses are low relative to
premiums, firms enter the market.
The relevant ratio that compares the overall expenses of the insurance industry with
premiums is:
Combined ratio= (total insurance expenses)/(net premiums)
When this ratio is low, i.e. less than 100%, new entrants join the market.
Investment returns
Property and casualty insurance companies should invest in steady-return, low risk, and
highly liquid assets.
When evaluating investments, the concentration of assets needs to be considered. Most of
the investments will be in fixed income instruments, therefore concentration by type,
maturity, credit quality, industry, or geographic location, etc. should be evaluated.
The investment performance can be measured as:
total investment income
Investment income =
invested assets
Liquidity
Given the nature of the property and casualty insurance business, high liquidity is essential.
Firms should be able to satisfy the claims as they arise.
To evaluate the liquidity of a company we can look at the hierarchy of fair value reporting.
The hierarchy is based on three levels:
 Level 1: The prices for these securities are readily available. They are traded in very
liquid markets.
 Level 2: These securities are traded in less liquid conditions. Their prices are
inferred from similar securities in liquid markets.
 Level 3: No pricing is explicitly available. The reported prices are based on models
and assumptions. This level represents the least liquid investments.

Capitalization
This refers to the amount of equity capital in a company. There are no global minimum
capital requirements. However, some local jurisdictions impose minimum requirements
based on the risk taken.
Key ratios used to analyze an insurance company are:
Ratio Formula What is Measured
Loss and loss Loss expense + loss adjustment expense Success in estimating
adjustment expense Net premiums earned risks insured
ratio
Underwriting expense Underwriting expense Efficiency of money
ratio Net premiums written spent in obtaining
new premiums
Combined ratio Loss and loss adjustment expense ratio Overall efficiency of
+ Underwriting expense ratio an underwriting
operation
Dividends to Diviends to policyholders (shareholders) Liquidity
policyholders Net premiums earned
(shareholders) ratio
Combined ratio after Combined ratio + Dividends to policyholders Efficiency of money
dividends (shareholders) ratio spent in obtaining
new premiums
Analyzing a life and health insurance company
Operations: Products and Distribution

There are three major sources of revenue for life and health insurance companies.
 Premiums
 Investment products and services
 Investment returns
L&H polices are longer term compared to P&C policies. Also, L&H claims are more
predictable compared to P&C claims. For these reasons, L&H companies can take more risk
and therefore earn a higher return on their investments.
Earnings characteristics
The major expense components for L&H companies are:
 Benefit payments
 Annuity contract payments
 Contract surrender expenses: Some insurance policies accumulate cash value and

policy holders have a right to cancel these policies before maturity. If cancelled, the
insurance company has to make a payment equal to the cash value.
When evaluating L&H companies, we should recognize that several expense items require
significant judgment and estimates. Also, the earnings can be distorted because the
accounting treatment across different items may vary.
We can use several ratios to evaluate the profitability for L&H companies:
 ROA, ROE, growth and volatility of capital, book value per share.
 Pre-tax and post-tax operating margins.
 Total benefits paid as a percentage of net premiums written and deposits. A lower
ratio is considered better.
 Commissions and expenses incurred as a percentage of net premiums written and
deposits. A lower ratio is considered better, it implies efficiency in selling insurance
contracts.
Investment returns, liquidity and capitalization
Investment returns:
L&H companies can take on more risk relative to P&C companies. Therefore, they can
generate a higher return on their investments. Evaluation of investment returns should
address:
 Diversification
 Investment performance
 Interest rate risk
Liquidity:
Liquidity requirement is driven by liabilities. If the liabilities are short term, then the
liquidity requirement is high. On the other hand, if liabilities are long term, then the
liquidity requirement is low.
Sources of liquidity include operating cash flows and sale of investment assets.
Liquidity is measured by comparing the amount of liquid assets relative to short term
liabilities. The insurance company should have sufficient liquid assets to cover the short-
term liabilities.
Capital:
There are no global requirements for capital like the Basel III. Instead, various jurisdictions
have their own capital adequacy standards based on risk profile. L&H companies need less
of an equity cushion compared to P&C companies because the payouts are more
predictable and less volatile.
Interest rate risk is a major factor in estimation of risk-based capital due to maturity
mismatch between assets and liabilities.

LM05 Evaluating Quality of Financial Reports

Quality of financial reports
Financial reporting quality

Low High
Low quality reports contain information High-quality reports contain information
that is pure fabrication. that is relevant, complete, neutral, and free
from error.
Low financial reporting quality impedes High financial reporting quality enables
assessment of earnings quality and impedes assessment.
valuation.
Earnings (results) quality

Low High
Low earnings quality is unsustainable. High earnings quality increases company
value.
High earnings quality is sustainable.
Quality spectrum of financial reports
From an investor’s perspective, the overall quality of financial reports, is the combination
of reporting quality and earnings quality, and it can be thought of as a continuous spectrum
ranging from highest to lowest as depicted in the diagram below:

Potential problems that affect the quality of financial reports
Income statement: Overstatement or non-sustainability of operating income and/or net

income.
 Overstated or accelerated revenue recognition
 Understated expenses
 Misclassification of revenue, gains, expenses, or losses
Balance sheet: Misstatement of balance sheet items
 Over- or understatement of assets
 Over- or understatement of liabilities
 Misclassification of assets and/or liabilities
Cash flow statement: Overstatement of cash flow from operations.
Mergers and acquisitions provide opportunities and motivations to manage financial
results. For example, companies with declining cash flow from operations (CFO) may
acquire companies to increase CFO. This is because the acquisition may appear in CFI if
paid with cash or not appear at all if paid for with equity.
General steps to evaluate the quality of financial reports
These two questions help an analyst in evaluating the quality of financial reports:
 Are the financial reports GAAP-compliant and decision-useful?

 Are the earnings of high quality? Do they provide an adequate level of return, and
are they sustainable?
Following are the general steps to evaluate the quality of financial reports:
1. Develop an understanding of the company’s business and its industry. Understand
what accounting principles a company and its competitors follow.
2. Learn about the management. Review disclosures about compensation and insider
transactions.
3. Identify significant accounting areas, especially those in which management
judgment or an unusual accounting rule is a significant determinant of reported
financial performance.
4. Make comparisons.
 Compare the company’s financial statements and significant disclosures in the
current year’s report with the financial statements and significant disclosures in
the prior year’s report.
 Compare the company’s accounting policies with those of its closest competitors
to see if there are any unusual revenue and expense recognition.
 Using ratio analysis, compare the company’s performance with that of its closest
competitors.
5. Check for warning signs of possible issues with the quality of the financial reports.
Examine the statement of cash flows with particular focus on differences between
net income and operating cash flows.
6. For firms operating in multiple segments by geography or product—particularly
multinational firms—consider whether inventory, sales, and expenses have been
shifted.
7. Use appropriate quantitative tools to assess the likelihood of misreporting.
Quantitative tools to assess the likelihood of misreporting
Beneish Model
The probability of manipulation is given by the M-score, which is estimated using a probit
model. The formula for calculating the M-score is given below:
M-score = –4.84 + 0.920 (DSR) + 0.528 (GMI) + 0.404 (AQI) + 0.892 (SGI) + 0.115 (DEPI) –
0.172 (SGAI) + 4.679 (Accruals) – 0.327 (LEVI)
where:
 M-score = Score indicating probability of earnings manipulation; normally
distributed random variable with a mean of 0 and a standard deviation of 1.
 DSR (days sales receivable index) = (Receivablest/Salest)/(Receivablest–1/Salest–1).
 GMI (gross margin index) = Gross margint–1/Gross margint.
 AQI (asset quality index) = [1 – (PPEt + CAt)/TAt]/[1 – (PPEt–1 + CAt–1)/TAt–1], where

PPE is property, plant, and equipment; CA is current assets; and TA is total assets.
 SGI (sales growth index) = Salest/Salest–1.
 DEPI (depreciation index) = Depreciation ratet–1/Depreciation ratet, where
Depreciation rate = Depreciation/(Depreciation + PPE).
 SGAI (sales, general, and administrative expenses index) = (SGAt /Salest)/(SGAt–
1/Salest–1).
 Accruals = (Income before extraordinary items – Cash from operations)/Total
assets.
 LEVI (leverage index) = Leveraget/Leveraget–1
How to interpret the M-score:
 The probability of earnings manipulation is calculated by using the cumulative
probabilities for a standard normal distribution.
 The probability is given by the amount on the left side of the distribution.
 Higher M-scores (less negative numbers) indicate an increased probability of
earnings manipulation.
Indicators of earnings quality
High earnings quality are the earnings that are sustainable and represent the returns equal
to or in excess of the company’s cost of capital. Low-quality earnings are insufficient to
cover the company’s cost of capital and/or are derived from non-recurring, one-off
activities.
Indicators of earnings quality include:
1. Recurring earnings
2. Earnings persistence and related measures of accruals
3. Mean reversion in earnings
4. Beating benchmarks
5. External indicators of poor-quality earnings
Evaluating the earnings quality of a company
The objective of analyzing earnings is to understand the persistence and sustainability of

earnings. Analysts must be aware of the following accounting choices used by managers to
enhance their companies’ reported performance:
Revenue Recognition:
 Understand the revenue recognition policies as stated in the most recent annual
report.
 Study the days’ sales outstanding (DSO) trend. If it is increasing, it means
receivables were not paid on time.

 Consider cash versus accrual revenue.

 Compare the DSO and receivables turnover with those of other companies from the
industry.
 Perform trend analysis, over time and in comparison with competitors, to assess
revenue quality.
 Look for transactions with related parties.
Property and Capital Expenditure Analysis:
 Understand the cost capitalization policies as stated in the most recent annual
report. This may be used to avoid expense recognition.
 Perform trend analysis to examine the following:
o Non-current asset accounts: may indicate an unusual increase in costs.
o Gross and operating profit margins
o Turnover ratio for total assets
o Depreciation compared to asset base
o Capital expenditure relative to gross PPE
 Look for company transactions with entities owned by senior officers or
shareholders.
Bankruptcy prediction models are used to quantify the likelihood that a company will
default on its debt and/or declare bankruptcy. One such model to assess the probability of
bankruptcy is the Altman model. Altman’s Z-score is calculated as follows:
Z-score = 1.2 (Net working capital/Total assets) + 1.4 (Retained earnings/Total assets) +
3.3 (EBIT/Total assets) + 0.6 (Market value of equity/Book value of liabilities) + 1.0
(Sales/Total assets)
The higher the Z-score, the lower the probability of bankruptcy of a company.
Evaluating cash flow quality
High quality operating cash flow means the company’s performance was good and the
company had high reporting quality.
High-quality cash flow for a mature company would typically mean that OCF:
1. is positive and derived from sustainable sources
2. is adequate to cover capital expenditures, dividends, and debt repayments, and
3. has low volatility.
Some OCF manipulating techniques used by companies are:
 Selling receivables to a third party
 Delaying payable to boost OCF
 Inappropriate classification of cash flows to overstate OCF

 Exploiting flexibility in accounting standards
Balance sheet quality
The characteristics of high financial reporting quality with regard to the balance sheet are:
 Completeness
 Unbiased measurement
 Clear presentation
The characteristics of high financial results quality with regard to a strong balance sheet
are:
 Optimal amount of leverage
 Adequate liquidity
 Economically successful asset allocation
Sources of information about risk
Some useful indicators of a company’s risk are:

 A company’s financial statements can provide useful indicators of financial,
operating, or other risk. For example, high leverage ratios signal financial risk.
 An audit opinion(s) covering financial statements provide information about
reporting risk.
 The notes and disclosures provide information about financial, operating, reporting,
or other risks.
 Required disclosures that are specific to an event, such as capital raising, non-timely
filing of financial reports, management changes, or mergers and acquisitions.
 Financial press, including online media, can be a useful source of information about
risk.

LM06 Integration of Financial Statement Analysis Techniques

Financial analysis framework
The financial statement analysis framework recommended by the CFA Institute consists of
the following steps:
1. Define the purpose and context of the analysis.
2. Collect input data.
3. Process input data, as required, into analytically useful data.
4. Analyze/interpret the data.
5. Develop and communicate conclusions and recommendations.
6. Follow up.
Integration of financial statement analysis techniques
DuPont analysis
 DuPont analysis decomposes ROE into its components:
1. ROE = Return on assets × Leverage
2. ROE = Net profit margin × Asset turnover × Leverage
3. ROE = EBIT margin × Tax burden × Interest burden × Asset turnover × Leverage
 Evaluating the different components of ROE allows the analyst to identify a company’s
potential strengths and weakness.
 DuPont analysis should be performed with and without the impact of ‘income from
associates’. This is because the operations and resources of associate companies are
not in the parent company’s control.
Asset base composition
Common size balance sheets are used to examine the change in asset base composition
over time.
Capital structure analysis
 An analyst can use leverage ratios and liquidity ratios to evaluate changes in the capital
structure of a company.
 However, simply looking at high-level leverage numbers is not enough. An analyst
should dive deeper to evaluate whether the capital structure is becoming riskier.
Segment analysis and capital allocation
 For large companies with multiple business segments it is important to perform a
segment analysis and understand how the company allocates funds to different
segments.
 The segment EBIT margin should be compared with the ratio of total capital
expenditure % to total assets.

 Ideally, segments with high margins should receive a relatively high capital allocation.
 However, high EBIT margins does not necessarily mean that a segment is doing well
from a cash-flow perspective. Beyond the EBIT margin, we should also estimate the
cash return on assets across different segments.
Accruals and earnings quality
 Earnings can be separated into cash flow and accruals. The accrual ratio can be
estimated using a balance sheet approach or a cash flow approach.
Balance sheet accruals ratio for time t = (NOAt - NOAt -1) / [(NOAt + NOAt-1)/2]
Cash flow accruals ratio for time t = [NIt - (CFOt + CFIt)] / [(NOAt + NOAt-1)/2]
 In both cases, if the accrual ratio is close to zero, this means that accruals play little or
no role in the financial performance of a company and the earnings are of high quality.
Cash flow relationships
 Cash flow from operations should be compared with operating profit to determine if the
earnings are of high quality.
 Other cash flow ratios such as the cash flow interest coverage ratio can also be used in
the analysis.
Decomposition and analysis of the company’s valuation
When a company has significant investments in associates, the price-to-earnings multiple
of the “pure company” should be calculated by removing the impact of associates and using
adjusted market value and net income.

LM07 Financial Statement Modeling

Top-down and bottom-up approaches
Top-down approach begins at the economy level, then the industry and finally to the
company level. There are two top-down approaches:
Growth relative to GDP growth: In this approach, we:
 Forecast nominal GDP growth rate (can forecast real GDP growth and inflation
separately).
 Forecast revenue growth relative to GDP depending on the company’s position in
the lifecycle and/or business cycle sensitivity. The forecasted revenue is expressed
in two ways:
o As percentage point discounts or premiums. For instance, Pfizer’s revenue is
projected to grow at 100 bps above nominal GDP growth rate.
o In relative terms: GDP is forecasted to grow at 4% and Oracle’s revenue is
forecasted to grow at a 25% faster rate.
Market growth and market share: In this approach, we:
 Forecast growth rate of relevant market
 Forecast change of company’s market share in the market. For example, assume
Tesla is expected to maintain a market share of 1% in the automobile market. If the
automobile market is expected to grow to $30 billion in annual revenue, then Tesla’s
annual revenue is forecasted to grow to 1% * $30 billion = $300 million.
Bottom-up approaches begin at the level of the individual company or unit within the
company. Examples of bottom-up approaches include:
 Time-series: Forecasts based on historical growth rates or time-series analysis.
 Return on capital: Forecasts based on balance sheet accounts and rates or ratios.
 Capacity-based measure: Forecasts (for example, in retailing) based on same-store
sales growth and sales related to new stores.
A hybrid approach combines top-down and bottom-up approaches.
Income statement modeling: Operating costs
Similar to revenue forecasting, costs may be forecasted using a top-down, bottom-up, or

hybrid approach
 Top-down approach: Consider factors such as overall level of inflation, or industry-
specific costs.
 Bottom-up approach: Consider factors such as segment-level margins, historical
cost-growth rates, historical margin levels, etc.
 Hybrid view: Incorporate elements from both top-down and bottom-up approaches.

A positive correlation between operating margins and sales suggests economies of scale.
Forecasting income statement elements
 Forecast COGS by segment, product category, or by volume and price, to improve
forecasting accuracy. COGS is forecasted as a percentage of sales using historical
relationships.
 In SG&A, selling and distribution costs such as wages are variable and can be
estimated as a percentage of sales. General and administrative expenses are more or
less fixed and increase or decrease gradually.
 Two non-operating expenses are financing expenses and taxes.
o To forecast debt financing expenses, forecast the level of debt and the
corresponding interest rates.
o There are three types of taxes: statutory tax rate, effective tax rate, and cash
tax rate. Use effective tax rate to forecast net income and the cash tax rate to
forecast cash flows. To forecast future tax expense, often the tax rate based
on normalized operating income, adjusted for special items, is used.
Balance sheet and cash flow statement modeling
The table below summarizes the approach for forecasting balance sheet items.
Item Forecasting method Comment
Accounts receivable Use sales and DSO Top-down: project economy
to slow sales down, lower
inventory turnover.
Bottom-up: use company’s
historical ratios
Inventory Use COGS and inventory Estimate maintenance and
turnover growth capital
expenditures.
PP&E and depreciation Consider the future need for
PP&E and depreciation
disclosures.
Debt (capital structure) Use debt/equity,
debt/capital, debt/EBITDA
Retained earnings Net income and dividend
policy
Cash flow statement can be derived based on the income statement and balance sheet.
Biases that influence analyst forecasts
Five key biases that influence analyst forecasts are:

 Overconfidence bias
 Illusion of control bias
 Conservatism bias
 Representative bias
 Confirmation bias
Impact of competitive factors in prices and costs
Competitive factors affect a company’s ability to negotiate lower input prices with
suppliers and to raise prices for products and services. Porter’s five forces framework can
be used as a basis for identifying such factors. The table below summarizes this framework.
Force Comment
Threat of substitutes Fewer substitutes or higher switching costs increase industry
profitability.
Internal rivalry Lower rivalry increases industry profitability.
Supplier power Presence of many suppliers limits their pricing power, which in
turn does not put a downward pressure on the industry’s
profitability.
Customer power Presence of many buyers limits their negotiating power, which
in turn does not put a downward pressure on the industry’s
profitability.
Threat of new entrants High barriers to entry increase industry profitability.
Sales and cost projections with inflation and deflation
Sales projections
 Higher input costs such as commodity or labor usually result in higher prices. This
relationship between price and cost, i.e. how long it takes for the price to increase, if
it’s gradual or immediate, depends on the industry structure.
 If demand is relatively price inelastic, revenue will benefit from inflation. If demand
is price elastic, revenues will decrease even if unit prices are increased because
volumes decline.
 High inflation in export market might increase sales in foreign currency terms, but
that gain might be lost if foreign currency depreciates.
Cost projections
The following factors help in forecasting costs:
 Understanding the purchasing characteristics of an industry. For example, if
companies use long-term price-fixed forward contracts and hedging instruments,
then price increases can be gradual instead of a sudden hike.
 Monitoring the underlying drivers of input prices such as weather conditions.

 The impact of inflation or deflation on industry’s cost structure depends on the

competitive structure.
 Break down the cost structure by category and geography.
 For each cost element, assess the impact of inflation and deflation on input prices;
consider the company’s ability to use cheaper alternatives for expensive inputs
when input prices rise.
Effect of technological developments on demand, selling prices, costs and margins
Technological developments can change the economics of industries and individual

companies. When a technological development results in a new product that threatens to
cannibalize demand for an existing product, a unit forecast for the new product combined
with an expected cannibalization factor can be used to estimate the impact on future
demand for the existing product.
The table below highlights the impact of technological developments on demand, selling
prices, costs and margins.
Impact on Factors to consider
Demand Market segments; cannibalization factor per segment. These help in
determining the post cannibalization volume.
Selling prices Company strategy; response to threat of substitutes
Operating costs Understand breakdown between fixed and variable costs. If sales
and margins decrease because of cannibalization, then variable costs also decrease.
Method 1: evaluate sensitivity of costs to changes in revenue
% Δ (cost of revenue + operating expense)
% Variable Cost ≈ % Δrevenue
% Fixed Cost = 1 − % Variable cost
Method 2: assign an estimate of the percentage of fixed and variable
costs to the various components of operating expenses
Long-term forecasting
Forecast time horizon is influenced by factors such as:

 Investment strategy for which the stock is being considered
 Cyclicality of the industry
 Company specific factors
 Analyst’s employer preferences
Longer-term projections often provide a better representation of the normalized earnings
potential of a company than a short-term forecast, especially when certain temporary
factors are present. Normalized earnings are the expected level of mid-cycle earnings for a
company in the absence of any unusual or temporary factors that impact profitability

(either positively or negatively).

“Growth relative to GDP growth” and “market growth and market share” methods can be
applied to develop longer term projections. Using revenue and cost forecasts, the income
statement, balance sheet and cash flow statement can be projected for the time horizon to
arrive a terminal value. The historical multiple-based approach is often used.

Corporate Issuers 2023 Level II High Yield Notes
LM01 Analysis of Dividends and Share Repurchases

Dividend types
Cash dividends are payments made to shareholders in cash. The three types of cash
dividends are:
 Regular cash dividends are paid out on a consistent basis. Stable or increasing
dividend is viewed as a sign of financial stability.
 Special dividends are one-time cash payments when the situations are favorable
(Also called as extra dividends or irregular dividends; used by cyclical firms).
 Liquidating dividend is distributed to shareholders when a company goes out of
business.
Due to cash dividends, the total shareholder wealth does not change. Only the distribution
between cash and share value changes. As cash is paid out of the company, the company’s
liquidity and leverage ratios become worse.
Stock dividends are payments made to shareholders in additional shares instead of cash.
Stock splits divide each existing share into multiple shares. Reverse stock splits are the
opposite of stock splits and decrease the total number of outstanding shares. Since stock
dividends and splits do not result in an outflow of cash, they do not affect the liquidity and
leverage ratios.
A share repurchase is when a company buys back its own outstanding shares using cash.
Due to stock dividends, there is no impact on shareholder’s wealth as there is no outflow of
cash. Also, the company’s leverage and liquidity ratios stay the same.
Divided policy theories

MM’s dividend irrelevance theory:
According to this theory, if we assume perfect capital markets (i.e., there are no taxes, no
transaction costs and symmetric information), then dividend policy does not matter. It will
have no impact on the cost of capital or on the company’s value.
There is no meaningful distinction between dividends and share repurchases. Investors can
generate homemade dividends. If investors want income, they can sell shares to create
dividends. Dividends can be used to purchase shares, if investors do not need income.
The bird in the hand argument:
This theory states that investors view cash dividends as less risky than the same amount of
capital gains. Hence, companies that pay dividends have a lower cost of equity as compared
to companies that do not pay dividends. The lower cost of equity results in a higher present
value for the companies that pay dividends.

MM counter argument: Whether or not dividends are paid, does not impact the riskiness of
future cash flows.
The tax argument:
This theory states that investors do not like dividends because dividends are taxed at a
higher rate than capital gains. Hence, they prefer companies that pay low dividends. This
increases the value of companies with low dividend payout ratios.
Signaling content of dividends

Usually, there is some level of information asymmetry, i.e., managers/ insiders know more
than outside investors. Hence, the board/management may use dividends to signal how the
company is really performing.
Dividend initiation or increase is generally seen as a positive signal and stock prices will
probably rise in the short-run. However, if revised dividends are not sustainable, stock
price will fall.
Decrease in dividends or omission of dividends is generally seen as a negative signal and
stock price will probably fall in the short-run.
Agency costs
A major concern to shareholders is that management might invest in negative NPV projects.
Such projects may increase the size of the company and thus be better for managers, but
they will generate negative economic returns.
This potential overinvestment problem can be partially controlled through high dividend
payouts. If most of the free cash flow is paid out, managers would be constrained in their
ability to overinvest by undertaking negative NPV projects.
However, dividend payments can worsen conflicts of interest between shareholders and
debt holders.
Factors affecting dividend policy

Factors that affect dividend policy are:
 Investment opportunities: All else equal, a company with many profitable
investment opportunities will pay out less dividends than a company with fewer
opportunities.
 The expected volatility of future earnings: When earnings are volatile, companies
are more cautious in the size and frequency of dividend increases.
 Financial flexibility: To obtain financial flexibility associated with holding more
cash, companies may not initiate, may reduce or omit dividends.
 Tax considerations: Under double taxation systems, dividends are taxed at both

the corporate and shareholder level. Under tax imputation systems, a shareholder
receives a credit on dividends for the tax paid on corporate profits. Under split-rate
taxation systems, corporate profits are taxed at different rates depending on
whether the profits are retained or paid out in dividends.
 Flotation costs: It is more expensive for companies to raise new equity capital than
to use their own retained earnings. As a result, many companies avoid establishing a
level of dividends that would create a need to raise new equity capital in future.
 Contractual and legal restrictions: Often the payment of dividends is affected by
legal or contractual restrictions or rules.
Effective tax rate

Double taxation:
Under double taxation systems, dividends are taxed at both the corporate and shareholder
level. The effective tax rate is given by,
Effective tax rate = Corporate tax rate + (1 – corporate tax rate) * (individual marginal tax
rate on dividends)
Tax imputation systems:
Under tax imputation systems, a shareholder receives a credit on dividends for the tax paid
on corporate profits. Corporate profits that are distributed as dividends are only taxed
once; the effective tax rate is equal to individual marginal tax rate of the investor.
Split-rate taxation systems:
Under split-rate taxation systems, corporate profits are taxed at different rates depending
on whether the profits are retained or paid out in dividends. Earnings distributed as
dividends are still taxed twice, but the comparatively low corporate tax rate on earnings
lessens that penalty.
Payout policies
Stable dividend policy:
In this policy, a company tries to align its dividend growth rate to the company’s long-term
earnings growth rate. The stable dividend policy can be represented by a gradual
adjustment process in which the expected increase in dividends is calculated as:
Expected increase in dividends = (Expected earnings × Target payout ratio – Previous
dividend) × Adjustment factor
Adjustment factor = 1/ number of years over which the adjustment in dividends will take
place
New dividend = Original dividend + Increase in dividend
Example: Current dividend = 0.4. Expected earnings for the year ahead = 1.50. Target

payout ratio = 0.5. Adjustment period = 5 years. Calculate the expected dividend.
Solution:
Adjustment factor = 1/5 = 0.2
Expected increase in dividends = (1.5 x 0.5 – 0.4) x 0.2 = 0.07
Expected dividend = 0.4 + 0.07 = 0.47
Constant dividend payout ratio policy:
In this policy, a company applies a target dividend payout ratio to current earnings;
therefore, dividends are more volatile than with a stable dividend policy.
Share repurchase methods

The four main types of repurchase methods are:
 Buy in the open market – has maximum flexibility because the company can time
the buyback. The method can be cost effective if share repurchase is timed to
minimize price impact and to exploit perceived undervaluation.
 Buy back a fixed number of shares at a fixed price – can be accomplished quickly
but the company usually has to offer a premium; pro-rata if necessary. Example:
stock selling at $37 and the company offers to buy 5 million shares for $40.
 Dutch auction – slower than the previous method, but can give a better price.
Example: stock selling at $37 and the company offers to buy 5 million shares in the
$38 to $40 range.
 Repurchase by direct negotiation – company buys back shares from a large
shareholder. This method helps avoid market impact of a large transaction.
Financial statement effects of repurchases

If net income remains unchanged, then a lower number of outstanding shares due to a
share repurchase can result in higher earnings per share. However, the change in net
income depends on how the repurchase is financed.
 If the share repurchase is financed internally, a repurchase increases EPS only if the
funds used for the repurchase would not earn their cost of capital if retained by the
company.
 If the repurchase is financed by taking external debt, the effect on EPS is positive if
the earnings yield exceeds the after-tax cost of financing the repurchase. If the
earnings yield is lower than the after-tax cost of financing, EPS would decrease.
If share repurchase is financed with surplus cash:
 EPS increases
 Assets and equity decrease
 Liquidity ratios decrease
 Debt to asset ratios increase

 Debt to equity ratios increase

If the market price per share is higher than book value per share, BVPS will decrease after
the share repurchase. On the other hand, if the market price per share is lower than book
value per share, BVPS will increase after the share repurchase.
Cash dividend vs. share repurchase decision

Theoretically, share repurchases are equivalent to cash dividends of equal amount in their
effect on shareholders’ wealth, all other things being equal. However, all other things are
not necessarily equal. Listed below are factors to consider when evaluating share
repurchases.
 Potential tax advantages
 Share price support/signaling that the company considers its shares a good
investment
 Added managerial flexibility
 Offsetting dilution from employee stock options
 Increasing financial leverage
Global trends in payout policy

The following broad trends have been observed:
 A lower percentage of companies in a given U.S. stock market index have paid
dividends than have companies in a comparable European stock market index.
 The fraction of companies paying cash dividends has been in long-term decline in
most developed markets (e.g., the United States, Canada, the European Union
overall, the United Kingdom, and Japan).
 Since the early 1980s in the United States and the early 1990s in the United
Kingdom and continental Europe, the fraction of companies engaging in share
repurchases has trended upward.
Analysis of dividend safety

Net Income:
The traditional measure of dividend safety is the dividend payout ratio and its inverse the
dividend coverage ratio.
Dividends
Dividend payout ratio =
Net Income
Net Income
Dividend coverage ratio =
Dividends
A higher dividend payout ratio or a lower dividend coverage ratio tends to indicate, all else
equal, higher risk of a dividend cut.
Free Cash Flow:

A more comprehensive measure of dividend safety is to use FCFE coverage ratio.

FCFE
FCFE coverage ratio =
(Dividends + Share repurchases)
If this ratio is 1, the company is returning all available cash to its shareholders. If the ratio
is significantly greater than 1, the company is building up its cash reserves. If the ratio is
significantly less than 1, the company is paying out more than it is earning. This is not
sustainable; it will deplete the cash reserves and the company will have to raise new equity.
An analyst should assess coverage ratios relative to historical levels and peer companies.
He should also look at the stage in the industry lifecycle the company is in while making his
decision.

LM02 ESG Considerations in Investment Analysis

Global variations in ownership structures and their effects on corporate governance
Corporate ownership structures can be classified as:
 Dispersed: Many shareholders exist and none of them have the ability to
individually exercise control over the company.
 Concentrated: An individual shareholder or a group has the ability to exercise
control over the company.
 Hybrid: Combination of the above two types.
Controlling shareholders may be either majority shareholders or minority shareholders.
Despite their low ownership stake, minority shareholders can still exercise control through
the following arrangements:
 Horizontal ownership: Companies have cross-holding share arrangements with each
other. They may also share key customers or suppliers. Such a structure can help
facilitate strategic alliances and foster long-term relationships among such
companies.
 Vertical (or pyramid) ownership: A company or group has a controlling interest in
two or more holding companies, which in turn have controlling interests in various
operating companies.
 Dual-class (or multiple-class) shares: One share class has superior or even sole
voting rights, whereas the other share class has inferior or no voting rights.
Different kinds of conflicts exist depending on the ownership structure and voting power.
1. Dispersed ownership and dispersed voting power: The managers may use the
company’s resources to pursue their own interests at the expense of shareholders.
This conflict is known as a principal-agent problem.
2. Concentrated ownership and concentrated voting power: The controlling owners
may be able to use the company’s resources to pursue their own benefit at the
expense of minority owners. This conflict is known as a principal-principal problem.
3. Dispersed ownership and concentrated voting power: This case is similar to case#2
and can lead to a principal-principal problem. The only difference is that the
controlling shareholders are minority shareholders and they do not own more than
50% of the company’s shares.
4. Concentrated ownership and dispersed voting power: This situation can arise when
there are legal restrictions on voting rights of majority shareholders, known as
voting caps. Voting caps are usually imposed by sovereign governments to prevent
foreign investors from exercising control over strategically important local
companies.
Types of influential owners include banks, families, sovereign governments, institutional
investors, group companies, private equity firms, foreign investors, managers, and board

directors.
Two typical structures for a company’s board of directors are:
 One-tier board - A one-tier board structure consists of a single board of directors,
composed of executive (internal) and non-executive (external) directors. Most
countries have a one-tier board structure.
 Two-tier board – A two-tier board structure consists of a supervisory board that
oversees a management board. Some countries mandate a two-tier board structure.
CEO duality exists when the chief executive officer (CEO) also serves as the chairperson of
the board.
Evaluating corporate governance policies and procedures

Board policies and practices: To evaluate a company’s board policies and practices, we
look at the following:
 Board of directors’ structure: Analysts consider whether the structure is a one-tier
or two-tier structure and if this structure provides sufficient oversight,
representation, and accountability to shareholders.
 Board independence: A board with a majority of independent board members is
considered positive.
 Board skills and experience: A board with diverse skills and expertise directly
related to the company’s core operations is considered positive.
 Board composition: A board with more diversity with respect to profession, culture,
geographical background, gender, age and tenure is considered positive.
An issue related to skills and experience is board tenure. A board member’s tenure
is considered long if it exceeds 10 years. This can be viewed positively (the board
member has more knowledge and experience) or negatively (the board member
could be too closely aligned with management).
Executive remuneration: Factors to consider while evaluating executive remuneration
include:
 Transparency of compensation
 Performance criteria for incentive plans – both short term and long term
 Linkage of remuneration with company strategy
 Pay differential between the CEO and the average worker
 “Say-on-pay” provision allows shareholders to vote and/or provide feedback on
remuneration issues.
 A “claw-back” policy allows a company to recover previously paid remuneration if
certain negative events are uncovered.
Shareholder voting rights:
 Under straight voting share structures, shareholders get one vote for each share

owned.
 Under dual-class share structures, company founders and management have shares
with more voting power.
Identifying and evaluating ESG-related risks and opportunities

Materiality and investment horizon:
There are several challenges while evaluating ESG information. Companies usually report
information and metrics that are inconsistent, making comparisons across companies
difficult. Also, ESG disclosures are voluntary. Therefore, different companies may report
different levels of information. This again makes comparisons across companies difficult.
While evaluating ESG factors, analysts first need to evaluate if an information is material.
Materiality typically refers to ESG related issues that can affect a company’s operation, its
financial performance, and the valuation of its securities.
Analysts also consider their investment horizon when deciding which ESG factors to
consider in their analysis. Some factors may affect a company’s performance in the short
term, whereas other issues may be more long term in nature.
Relevant ESG-related factors:
Analysts typically use three approaches to identify a company’s (or industry’s) ESG factors:
1. Proprietary research: involves the use of own judgment or their firm’s proprietary
tools to identify ESG information.
2. ESG data providers: involves the use of information supplied by ESG data providers,
such as MSCI or Sustainalytics.
3. Not-for-profit industry organizations and initiatives: involves the consideration of
not-for-profit initiatives that provide data and insights on ESG issues.
Equity vs. fixed-income security analysis
 Equity analysis: ESG integration is used to both identify potential opportunities and
mitigate downside risk. Analysts adjust financial model variables such as cost of
capital to reflect ESG factors.
 Fixed-income analysis: ESG integration is generally focused on mitigating downside
risk. Analysts usually do not focus on potential opportunities. The credit assessment
may vary depending on maturity.
ESG integration
The typical starting point for ESG integration is the identification of material qualitative
and quantitative ESG factors that pertain to a company or its industry.
For equities, adjustments related to ESG factors may be made to cost of capital, price
multiple, and terminal value.
For bonds, an analyst may adjust an issuer’s credit spread or CDS to reflect anticipated
effects from ESG considerations.

LM03 Cost of Capital - Advanced Topics

Cost of capital factors
Several factors can influence a company’s cost of capital. These factors can be either top-
down (i.e., external or systematic) or bottom-up (i.e., company specific or idiosyncratic).
These factors are listed in the table below.
Top-Down, External Bottom-Up, Company Specific
Capital availability Revenue, earnings, and cash flow volatility
Market conditions (risk free rates, credit Asset nature and liquidity
spreads, and ERP)
Legal and regulatory Financial strength, profitability, and
considerations/country risk leverage
Tax jurisdiction Security features (Features of debt
securities, such as callability, putability, and
convertibility, affect the cost of debt.
Features of equity securities, such as
cumulative dividends, affect the costs of
equity)
Estimating the cost of debt

The cost of debt varies depending on the type of debt, the liquidity of the debt issue, the
credit rating of the debt, and the currency in which the debt is issued.
If a company has publicly traded debt with no embedded options, then the YTM on the
company’s existing debt with the longest maturity can be interpreted as the current cost of
issuing new debt.
For non-traded debt, the analyst needs to use approaches like synthetic credit rating and
matrix pricing.
Determining the cost of bank debt and leasing requires information for the calculation of
the effective cost of this financing.
The cost of debt in international markets can be estimated using country risk rating (CRRs)
that reflect economic, political, and exchange rate risk, as well as information about the
financial markets and regulation.
The ERP (Equity Risk Premium)

The equity risk premium is the incremental return that investors require for holding
equities rather than a risk-free asset.
There are two broad approaches to estimate the equity risk premium (ERP): historical

estimates and forward-looking estimates.

A historical equity risk premium estimate is calculated as the mean value of the differences
between broad-based equity-market-index returns and government debt returns over a
given time period.
Forward-looking estimates (or ex-ante estimates) are based on current information and
expectations concerning economic and financial variables. The three main categories of
forward-looking estimates are:
 Survey estimates: Ask experts for their opinion.
 Dividend discount models: ERP = market dividend-yield + consensus long-term
earnings growth rate – current long-term government bond yield
 Macroeconomic modeling: ERP = [DY + Δ(P/E) + i + g + ΔS] – E(rf)
Estimating the required return on equity
DDMs: The stock’s required rate of return is composed of two components: dividend yield
D
( P1 ) and price appreciation (g).
0
D1
re = +g
P0
Bond Yield Plus Risk Premium Approach:
Cost of equity = YTM on the company’s long-term debt + risk premium
The YTM on the company’s long-term debt already includes: the real interest rate, a
premium for expected inflation, and a default risk premium. The additional risk premium in
the formula compensates for the added risk arising from holding the firm’s equity.
Risk-Based Models
CAPM:
According to CAPM, the required return on an asset can be calculated as:
re = rf + β̂ (ERP)
Fama–French Models:
Using the three-factor model, a company’s excess return on equity is calculated as:
re = rf + β1 ERP + β2 SMB + β3 HML
where:
SMB is the size premium, equal to the average difference in equity returns between
companies with small and large capitalizations
HML is the value premium, equal to the average difference in equity returns between

companies with high and low book-to-market ratios

The five-factor Fama–French model adds two other factors—a profitability factor (RMW)
and an investment factor (CMA):
re = rf + β1 ERP + β2 SMB + β3 HML + β4 RMW + β5 CMA
where:
RMW is the profitability premium, equal to the average difference in equity returns
between companies with robust and weak profitability
CMA is the investment premium, equal to the average difference in equity returns
between companies with conservative and aggressive investment portfolios
Estimating the Cost of Equity for Private Companies:
The required return on equity for private companies often includes:
 a size premium (SP),
 an industry risk premium (IP), and
 a specific-company risk premium (SCRP)
Two commonly used methods to estimate the required return on equity for private
companies are:
 the expanded CAPM and
 the build-up approach
Expanded CAPM:
re = rf + βpeer (ERP) + SP + IP + SCRP
Build-Up Approach:
re = rf + ERP + SP + SCRP
International Considerations:
Investing in emerging markets exposes investors to additional risks as compared to
investing in developed markets. Two methods for building risk premiums into the required
return for emerging markets are:
 the country spread model
 the country risk rating model
Country spread model:
According to the country spread model, the ERP for an emerging equity market can be
calculated as:
ERP = ERP for a developed market + (λ × Country risk premium)

where:
λ is the level of exposure of the company to the local country
Country risk premium = Sovereign yield spread x σEquity/ σBond
International CAPM:
Under the international CAPM (ICAPM) approach, the returns on a stock in an emerging
market are regressed against the risk premium of a global index (rgm) in addition to that of
wealth-weighted foreign currency index ( rc ):
E (re) = rf + βG (E(rgm) − rf) + βC (E(rc) − rf)

LM04 Corporate Restructuring

Types of corporate restructuring and motivations for pursuing them
Corporate restructuring can be classified as one of nine types as shown in Exhibit 5 from
the curriculum.
Exhibit 3 from the curriculum lists the typical motivations for corporate structural change.
Investment Actions Divestment Actions Restructuring
Actions
Issuer-specific • Realize synergies • Focus operations • Improve returns on
motivations • Increase growth and business lines capital
• Improve • Valuation • Financial challenges,
capabilities or secure • Liquidity needs including bankruptcy
resources • Regulatory and liquidation
• Opportunity to requirements
acquire an
undervalued target
Top-down • High security prices
drivers • Industry shocks

Evaluating corporate restuructring

The evaluation of corporate restructurings is composed of four phases as shown in Exhibit
12 from the curriculum.
The initial evaluation of a corporate restructuring involves answering four questions:

 What is happening?
 Why is it happening?
 Is it material?
 When is it happening?
Preliminary valuation
If a restructuring is material, an analyst should conduct a preliminary valuation of the
target to check if the management is using stakeholder resources optimally. Three
valuation methods commonly used in this step are: comparable company analysis,
comparable transaction analysis, and premium paid analysis.
Comparable Company Analysis
The steps for the comparable company approach are:
1. Define a set of comparable firms. Include companies in similar industries that have
similar size and capital structure.
2. Calculate various relative value measures based on the current market price of the
comparable companies. For example, EV/EBITDA, P/S, P/E, P/CF, etc.
3. Review descriptive statistics like mean, median, and range for the metrics chosen
and apply judgment and experience to estimate company value.
The comparable company analysis is often used to assess the value of targets in spin offs

rather than acquisitions. This is because in acquisitions, acquirers pay a premium for
control and the acquisition price typically exceeds the price estimated by this method.
Comparable Transaction Analysis
The steps for the comparable transaction approach are:
1. Collect a sample of recent and relevant takeover transactions.
2. Calculate various relative measures like P/E, P/CF, P/B based on prices actually paid
in the deals (not current market prices).
3. Review descriptive statistics like mean, median, and range for the metrics chosen
and apply judgment and experience to estimate company value.
Unlike comparable company analysis, the valuation multiples in comparable transaction
analysis include takeover premiums, because they reflect historical acquisitions.
Premium Paid Analysis
To estimate or judge a sale value or acquisition price for a listed issuer, an analyst could
also calculate an estimated takeover premium.
The takeover premium is the amount by which the takeover price for each share of stock
must exceed the current stock price in order to entice shareholders to relinquish control of
the company to an acquirer. This premium is usually expressed as a percentage of the stock
price and is calculated as:
(DP − SP)
PRM =
SP
where:
PRM = takeover premium (as a percentage of stock price)
DP = deal price per share of the target company
SP = stock price of the target company
To calculate a relevant takeover premium, analysts compile a list of takeover premiums
paid for companies similar to the target in the recent past.
Modeling and valuation

The next step of the evaluation process is creating pro forma financial statements that
include the effect of the restructuring.
Exhibit 15 from the curriculum demonstrates how to create a pro forma income statement.
Pro Forma Income Statement (Acquisition) Modeling
Revenue 1. Combine acquirer and target revenues.
2. Add revenue synergies or subtract the cost of incompatible
activities (dis-synergies).

Operating expenses 1. Combine acquirer and target operating expenses.

2. Subtract cost synergies or add the cost of incompatible
activities (dis-synergies).
Depreciation and 1. Combine acquirer and target depreciation and amortization.
amortization 2. Add amortization of acquired intangible assets.
Other expense or 1. Combine acquirer and target other expense or income.
income
Interest expense 1. Start with current acquirer interest expense.
2. Add increased interest from new debt issuance and revised
interest rate.
Income taxes 1. EBT-weighted average of tax rates of acquirer and target;
estimate usually provided by issuer
Shares outstanding 1. Start with current acquirer shares outstanding
2. Add shares from any share issuance
Pro forma WACC
The WACC for an issuer is determined by the weights of different capital types and the
constituent costs of capital. A restructuring can change both the weights and the costs of
each type of capital. The costs of capital are influenced by several factors both inside and
outside the issuer. Restructurings change the costs of capital by changing these factors.

Equity 2023 Level II High Yield Notes
LM01 Equity Valuation: Applications and Processes

Intrinsic value and sources of perceived mispricing
Intrinsic value is the true or real value given a hypothetically complete understanding of
the asset’s investment characteristics. Intrinsic value might be different from the current
(market) price of an asset.
Perceived mispricing is the difference between the estimated intrinsic value and the
market price of an asset. It can be expressed as:
VE – P = (V – P) + (VE – V)
Interpretation:
 The first component (V-P) is the true mispricing or market error. It is the difference
between the unobservable intrinsic value V, and the observed market price P. An
analyst exploits the mispricing if it arises from the first term as it contributes to the
abnormal return.
 The second component (VE – V) is the error in the estimate of the intrinsic value or
analyst error. It is the difference between the valuation estimate and the
unobservable intrinsic value. The smaller the second term, the better it is for the
analyst.
For active investing to be successful, an analyst’s estimates must be different from the
consensus estimate and must be correct.
Going-concern value and liquidation value
A going-concern assumption is the assumption that the company will continue its business
activities into the foreseeable future, i.e. it will continue to produce and sell its products
and services. The value of a company based on this assumption is called the going-concern
value.
On the other hand, liquidation value is the value of a company if it were dissolved and its
assets sold individually.
Fair market value and investment value
Fair market value is the price at which an asset (or liability) would change hands between
a willing buyer and a willing seller when the former is not under any compulsion to buy
and the latter is not under any compulsion to sell.
Fair value is the price of an asset used for financial reporting, for example, impairment
testing.
Investment value is the value to a specific buyer based on potential synergies and the

investor’s requirements and expectations.

Note: Of the many definitions of value, the definition most relevant to public company
valuation is intrinsic value calculated using the going-concern assumption.
Applications of equity valuation
Analysts use valuation concepts and models in the following practical scenarios:
 Selecting stocks
 Inferring market expectations
 Evaluating corporate events
 Rendering fairness opinion
 Evaluating business strategies
 Appraising private businesses
Industry and competitive analysis
The five steps in the valuation process are:

1. Understanding the business
2. Forecasting company performance
3. Selecting the appropriate valuation model
4. Converting forecasts to a valuation
5. Applying the valuation conclusions
The questions that should be addressed in conducting an industry and competitive analysis
include:
 Evaluate industry prospects: how attractive are the industries in which the company
operates, in terms of offering prospects for sustained profitability? Use Porter’s five
forces framework to understand industry structure.
 Perform a competitive analysis: what is the company’s relative competitive position
within its industry, and what is its competitive strategy - cost leadership or
differentiation?
 Evaluate execution of the strategy: how well has the company executed its strategy
and what are its prospects for future execution? Analyze the past and current
financial reports to assess how successful is the strategy.
Absolute and relative valuation models
An absolute valuation model estimates an asset’s intrinsic value that can be compared to
the asset’s market price. The two types of absolute valuation models are:
 Present value model: The value of an asset is the present value of its expected future
cash flows.
 Asset-based valuation: Value of a company is equal to the market value of the assets

or resources it controls.
A relative valuation model estimates an asset’s value relative to that of another asset. The
underlying notion is that similar assets should sell at similar prices. Commonly used
multiples include:
 Price multiples: Ratio of a stock’s price to a fundamental such as earnings, book
value or cash flow per share.
 Enterprise multiples: Ratio of the total value of a firm to a fundamental such as
EBITDA.
Sum-of-the-parts valuation and conglomerate discounts
A valuation that sums the estimated values of each of the company’s businesses as if each
business were an independent going concern is known as a sum-of-the-parts valuation.
A conglomerate discount is a discount applied to the stock of a company operating in
multiple, unrelated businesses compared to the stock of companies with narrower focuses.
It is sometimes applied to value companies that require a sum-of-the-parts valuation.
Choosing a valuation approach
The criteria for choosing an appropriate approach for valuing a given company implies that
the model is:
 consistent with the characteristics of the company
 appropriate given the availability and quality of data
 consistent with the purpose of valuation or analyst’s perspective
Two important aspects of converting forecasts to valuation are:
 Sensitivity analysis that determines how changes to an input variable changes the
value of equity.
 Situation adjustments such as control premium, lack of marketability discount,
illiquidity discount, blockage factor, etc.

LM02 Discounted Dividend Valuation

Streams of expected cash flows
Discounted cash flow models are based on the idea that value of a security today is equal to
the present value of its future cash flows.
The three most widely used definitions of cash flows are dividends, free cash flow (FCFF
and FCFE), and residual income.
The suitability of each of these models is summarized in the table below:
Dividend discount model Free cash flow model Residual income model
The company has a history Company pays no dividends. Company pays no dividends.
of dividend payments. Or the dividends
significantly exceed or fall
short of free cash flow to
equity.
The investor takes a non- The investor takes a control
control perspective. perspective.
The board of directors has The company’s free cash The company’s expected free
established a dividend policy flows align with the cash flows are negative
that bears an company’s profitability within the analyst’s
understandable and within a forecast horizon comfortable forecast
consistent relationship to with which the analyst is horizon.
the company’s profitability. comfortable.
Applicable to mature, The firm has accounting
profitable firms disclosures of high quality.
Dividend discount model (DDM)
The value of a stock using DDM for a single-holding period is:

D1 P1 D1 + P1
V0 = + =
(1 + r)1 (1 + r)1 (1 + r)1
The value of a stock using DDM for multiple finite holding periods is:
n
Dt Pn
V0 = ∑ +
(1 + r) t (1 + r)n
t=1
The value of a stock for infinite holding periods is:

𝛼
Dt
V0 = ∑
(1 + r)t
t=1

There are two approaches to forecast dividends.

 One is to assume that they follow a stylized growth pattern (constant growth, two-
stages of growth, or three stages of growth).
 The other alternative is to forecast dividends for a finite period, then forecast the
remaining dividends based on a pattern or by calculating the terminal stock price.
Gordon growth model
The Gordon growth model assumes that:

 Dividends grow at a constant growth rate g.
 Discount rate r is constant and is greater than g.
 Dividends bear and understandable and consistent relationship with profits.
The value of a stock using the Gordon growth model is:
D1
V0 =
r − g
If prices are efficient (price equals value), the price is expected to grow at a rate of g,
known as the rate of price appreciation. In this case, the stock’s expected rate of return is:
D1
r= +g
P0
If an estimate of the next-period dividend and the stock’s required rate of return are given,
then the Gordon growth model can be used to estimate the dividend growth rate implied by
the current market price.
D1
g= r−
P0
Present value of growth opportunities (PVGO)
The present value of growth opportunities (PVGO), also known as the value of growth, is
the present value of opportunities to profitably reinvest future earnings.
A stock’s value comprises of two parts:
 present value of the company with no growth component and
 present value of growth opportunities.
E1
V0 = + PVGO
r
E1
If price = value, then PVGO = P0 − r

Justified leading and trailing P/E ratios
The Gordon growth model can also be used to estimate justified leading and trailing P/E
ratios based on the fundamentals of the firm.
The leading P/E ratio is:
D1
P0 E 1−b
= 1 =
E1 r − g r − g
The trailing P/E ratio is:

D1 D0
P0 (1 + g) (1 − b)(1 + g)
E0 E0
= = =
E0 r − g r−g r−g
Value of a perpetual preferred stock
The value of a non-callable, fixed-rate, perpetual preferred stock is:

D
V0 =
r
Strengths and limitations of the Gordon growth model
The strengths of the Gordon growth model are:

 It is the simplest practical implementation of the dividend discount model.
 Broad equity market indices of developed markets frequently satisfy the conditions
of the model fairly well. It is used to judge whether the equity market is fairly valued
or not.
 It is applicable to stable, mature, dividend-paying firms that have a constant growth
in dividends. It can also be applied to firms that repurchase stock.
 g can be estimated using macroeconomic data; nominal GDP = real GDP + long-term
inflation.
 It is most suitable for a single stage DDM or can be used to model the last stable
stage in a multi-stage DDM.
The limitations of the Gordon growth model are:
 The model cannot be used if growth rate g is not constant. Since most firms have
non-constant growth in dividends, the model is not reliable. A multi-stage growth
model is more appropriate.
 The model cannot be used to value non-dividend paying firms. Dividends must also
be related to the level of earnings.
 Any small changes to (r – g) estimates can have a significant impact on stock value.

Multistage dividend discount models
Companies with varying growth prospects must be valued using multistage DCF models
such as the two-stage DDM, H-model, three-stage DDM or spreadsheet modeling. The
assumptions of these models are:
 Two-stage DDM: this model assumes different growth rates in stage 1 and stage 2.
A supernormal growth in stage 1 is followed by a lower, sustainable growth rate in
second stage.
 H-model: the dividend growth rate declines linearly from a high supernormal rate
to the normal growth rate in stage 1, and becomes constant and normal in stage 2.
 Three-stage DDM: There are two variations of this model.
o In one version, the growth in the middle stage is constant.
o In the second version, the middle stage is similar to the first stage in the H-
model. Dividends grow at a constant high rate in the second stage, and
declines linearly to a sustainable, mature growth rate in the third stage. The
second and third stages can be valued as an H-model.
The value of a stock using two-stage DDM is:
N
D0 (1 + g s )t D0 (1 + g s )n (1 + g L )
V0 = ∑ +
(1 + r)t (1 + r)n (r − g L )
t=1
where:
gs = short-term high growth rate
gL = long-term sustainable growth rate
n = number of years of high growth rate
The value of a stock using the H-model is:

D0 (1 + g L ) + D0 ∗ H ∗ (g s − g L )
V0 =
r − gL
where:
H = half-life in years of the high-growth period; high-growth period = 2H years
gS = initial short-term dividend growth rate
gL = normal long-term dividend growth rate after year 2H
The strength of multistage models is that they are flexible to pattern different growth
patterns in dividends.
However, the multistage models have several limitations:
 Often, the present value of the terminal stage represents more than three-quarters
of the total value of shares.
 Terminal value can be very sensitive to the growth and required return

assumptions.
 Technological innovation can make the lifecycle model a crude representation.
Phases of growth
Most companies experience varying phases of growth. The growth of most publicly traded
companies can be classified into the following three categories of growth:
 Growth phase: Companies in a growth phase enjoy
o Rapidly expanding markets
o High profit margins
o Abnormally high growth rate in earnings per share
On the flip side, they have:
o Negative free cash flows to equity
o Low dividend payouts
o Earnings growth rates may decline eventually as they attract new
competitors
 Transition phase: for companies in transition phase:
o Earnings are still rising, but at a slower pace as there is pressure on prices
and profit margins. The earnings growth rate is above average but is moving
towards the nominal growth rate of the economy.
o There is a decline in capital requirements, which leads to positive free cash
flow.
o Increasing dividend payout ratios
 Mature phase: for companies in mature phase:
o Return on equity equals required return on equity
o Earnings growth, dividend payout ratio, and return on equity stabilize at the
mature growth rate
o Can be valued using the Gordon growth model
Determining terminal value
The terminal value of a stock can be found using two methods:

 the Gordon growth model or
 multiplying the forecasted multiple such as P/E by forecasted EPS as of the terminal
date.
Determining a required return
A required return can be calculated if all the inputs to a DDM and price are given. For a
Gordon growth model, the return can be calculated using the formula:

D1
r= +g
P0
Formula for calculating expected return for the H-model:

D0
r = ( ) [(1 + g L ) + H(g S − g L )] + g L
P0
For multistage models and spreadsheet models, required return is calculated through trial
and error by equating the present value of cash flows to the current stock price.
Spreadsheet modeling
 The DDM models, such as the Gordon growth model or the H-model, assume
patterns of growth. Spreadsheet modeling makes it easier to model any growth
pattern.
 It allows the analyst to build complicated models that would be very cumbersome to
describe using Algebra, or where an iterative process is necessary.
 Built-in spreadsheet functions (such as those for finding rates of return) use
algorithms to get a numerical answer when a mathematical solution would be
impossible or extremely challenging.
 Several analysts can work together or exchange information by sharing their
spreadsheet models.
Sustainable growth rate of a company
The sustainable growth rate is the dividend/earnings growth rate that can be sustained for
a given level of return on equity, assuming that the capital structure stays constant through
time and the firm issues no new stock.
The formula for calculating the sustainable growth rate is: g = b ∗ ROE
The sustainable growth rate using the DuPont analysis is:
Net Income − Dividends Net Income Sales Total Assets
g= ∗ ∗ ∗
Net Income Sales Total Assets Shareholder ′ s Equity
An easy way to remember this formula is:
g=P*R*A*T
where, P = profit margin; R = retention rate; A = asset turnover; T = financial leverage
Valuing a stock using DDM
 If the DDM estimate of a stock’s value is more than its current price, then the stock is
undervalued.
 If the DDM estimate of a stock’s value is less than its current price, then the stock is

overvalued.
 If the DDM estimate of a stock’s value is equal to its current price, then the stock is
fairly valued.

LM03 Free Cash Flow Valuation

FCFF and FCFE approaches to valuation
Free cash flow to the firm (FCFF) is the cash flow available to the company’s suppliers of
capital after all operating expenses (including taxes) have been paid and necessary
investments in working capital and fixed capital have been made. A firm’s suppliers of
capital include common stockholders, bondholders, and preferred stockholders.
Free cash flow to equity (FCFE) is the cash flow available to the company’s common
stockholders after all operating expenses, interest, and principal payments have been paid
and necessary investments in working and fixed capital have been made.
The two methods to value equity using free cash flows are:
 Discount FCFF at the weighted average cost of capital (WACC) because FCFF is an
after-tax cash flow. Then, estimate the value of equity by subtracting the value of
debt from the estimated value of the firm.
 Discount FCFE at the required return of equity.
FCFF is preferred over FCFE for
 a levered company with negative FCFE, or
 for a company with changing capital structure.
In free cash valuation, the focus is on the value of assets needed to generate operating cash
flows. Analysts often exclude non-operating assets such as cash and marketable securities.
However, if the non-operating assets are significant and were excluded, then they must be
added to the value of operating assets.
Value of firm = Value of operating assets + Values of non-operating assets
Ownership perspective implicit in the FCFE approach
Analysts prefer to use the FCFF or FCFE approaches if one of the following conditions exist:
 The company does not pay dividends.
 The company pays dividends but the dividends do not reflect the company’s
capacity to pay dividends.
 The investor takes a control perspective.
Free cash flow approaches reflect a control (ownership) perspective, whereas DDM
approaches reflect a non-control perspective. An ownership perspective implies that the
acquirer can take control of the company and change the dividends substantially.
Therefore, free cash flow approaches make more sense than DDM approaches.

Calculating FCFF and FCFE
FCFF
FCFF from NI: FCFF = NI + NCC + Int (1 − tax rate) − FCInv − WCInv
FCFF from EBIT: FCFF = EBIT (1 – Tax rate) + Dep – FCInv – WCInv
FCFF from EBITDA: FCFF = EBITDA (1 – Tax rate) + Dep (Tax rate) – FCInv – WCInv
FCFF from CFO: FCFF = CFO + Int (1 – Tax rate) - FCInv
FCFE
FCFE from FCFF: FCFE = FCFF – Int (1 – Tax rate) + Net Borrowing
FCFE from NI: FCFE = NI + NCC – FCInv – WCInv + Net borrowing
FCFE from CFO: FCFE = CFO – FCInv + Net borrowing
To determine the WCInv, we ignore cash and short-term debt. For example, if we are given
the following information about a company:
Y1 Y2
Cash 10 12
AR 20 22
Inv 30 33
AP 10 10
Short-term debt 14 17
WCInv = (22 + 33 – 10) – (20 + 30 – 10) = 5
Forecasting FCFF and FCFE
There are two approaches to forecast FCFF and FCFE:

 Constant growth: One approach is to assume that the free cash flows (FCFF and
FCFE) grow at a constant rate. The simplest assumption is to use the historical
growth rate if the relationships between free cash flow and the fundamental factors
are expected to continue.
 Forecast individual components: The second approach is to forecast the
individual components of free cash flow, such as EBIT, net noncash charges,
investment in fixed capital, and investment in working capital.
The equation for FCFE assuming the Debt ratio (DR) is maintained is:
FCFE = NI - (1 – DR) (FCinv – Dep) – (1 – DR) (WCInv)
FCFE model versus dividend discount models

Analysts prefer the FCFE model over dividend discount model for the following reasons:
 Some companies pay no or low dividends.
 Dividends are substantially lower or more than their free cash flow.
 Free cash flow to equity is the amount available for distribution without impairing a
company’s value.
 FCFE is an appropriate cash flow measure when a company is a target for a
takeover.
Impact of dividends, share repurchases, share issues, and changes in leverage on

FCFF and FCFE
Dividends, share repurchases, and share issues do not affect FCFE and FCFF.
Increase in leverage increases the tax savings for FCFF.
Leverage affects FCFE in two ways:
 When new debt is issued, FCFE increases by the amount of debt issued.
 Later, FCFE decreases as interest expense increases.
Using net income and EBITDA as proxies for FCFE and FCFF
EBITDA is a poor proxy for cash flow because it does not account for depreciation,
investments in fixed capital and working capital, and cash outflow due to taxes.
FCFF = EBITDA (1 – Tax rate) + Dep (Tax rate) – FCInv – WCInv
Similarly, net income is a poor proxy for FCFE because it does not account for fixed capital,
working capital investment and net borrowings.
FCFE = NI + NCC – FCInv – WcInv + Net borrowing
Estimating company value using cash flow models
Single-stage (constant-growth) FCFF and FCFE models:

FCFF1
Firm value =
WACC − g
FCFE1
Equity value =
r − g
Multi-stage FCFF and FCFE models

There are several versions of the multi-stage model. In one version, we estimate the free
cash flows up to a certain number of years and assume that the free cash flows will grow at
a constant rate from there on. The formulas for this version are:

n
FCFFt FCFFn+1 1
Firm value = ∑ + ∗
(1 + WACC)t WACC − g (1 + WACC)n
t=1
n
FCFEt FCFEn+1 1
Equity value = ∑ + ∗
(1 + r) t r−g (1 + r)n
t=1
Another version assumes declining growth in Stage 1 followed by a long-run sustainable

growth rate in Stage 2. We can use the H-model formula (discussed in DDM) for this
version.
Three-stage growth models are appropriate for companies that have three distinct stages
of growth – growth phase, transition phase and mature phase.
Sensitivity analysis in FCFF and FCFE valuations
The value of a firm or equity depends on estimates for future growth rates, duration of
growth, and base-year values for FCFF and FCFE. Growth rate and duration of growth
depend on the growth phase of the company and the profitability of the industry.
Valuation models are sensitive to these inputs as the values can have a large impact on the
value of equity or firm. Analysts perform a sensitivity analysis to examine how valuation
changes with each of these inputs.
Calculating the terminal value in a multistage valuation model
The terminal value is calculated using two ways as seen in DDM:

 a single-stage model which assumes constant growth in the cash flows forever.
 a multiples approach which uses a valuation multiple such as the P/E ratio.
Valuing a stock using free cash flow valuation model
 If the model’s estimate of a stock’s value is more than its current price, then the
stock is undervalued.
 If the model’s estimate of a stock’s value is less than its current price, then the stock
is overvalued.
 If the model’s estimate of a stock’s value is equal to its current price, then the stock
is fairly valued.

LM04 Market-Based Valuation: Price and Enterprise Value Multiples

Price multiples in valuation
Price multiples are ratios of a stock’s market price to some measure of fundamental value
per share, for example – price to earnings (P/E) ratio. There are two methods to use price
multiples: method of comparables and method based on forecasted fundamentals.
 The method of comparables compares the price multiple of an asset with
comparable or similar assets. The price multiple of an asset is compared to that of a
benchmark value of the multiple to evaluate if it is overvalued, undervalued, or
fairly valued. The benchmark value of the multiple may be the mean or median
value of a peer group of companies, an industry or sector, an equity index, or the
historical price multiple of the stock. The economic rationale for the method of
comparables is the law of one price: identical assets should sell at the same price.
 The method based on forecasted fundamentals compares the actual value of a
firm or its equity with one based on its forecasted fundamentals. The economic
rationale behind this method is that fundamentals drive future cash flows, and
multiples can be related to fundamentals through its DCF value.
The justified price multiple is an estimate of the fair value of a multiple using either the
method of comparables or the method of forecasted fundamentals. It represents what the
price multiple should be if the stock is fairly valued.
Commonly used price multiples
Price to earnings (P/E): Ratio of a stock’s market price to its earnings per share.
There are two variations of P/E: the trailing P/E and the forward P/E.
 Trailing P/E is a stock’s current market price divided by the most recent four
quarters’ EPS.
 The forward P/E or leading P/E, is a stock’s current price divided by next year’s
expected earnings.
Rationales for using P/E include:
 Earnings are the key driver of value.
 P/E ratio is widely recognized and used by investors.
 Differences in P/Es may be related to differences in long-run average returns.
Drawbacks of using P/E include:
 Earnings can be negative which will make the ratio meaningless.
 Earnings often contain volatile, transient components.
 Management’s choice of accounting policies may taint EPS’s value.

Price to book value (P/B): Price divided by book value per share.
Book value per share represents, on a per-share basis, the investment that common
shareholders have made in the company. Book value will generally need to be adjusted.
When comparing, we often use tangible book value.
Rationales for using P/B include:
 Can be used when EPS is negative
 BVPS is more stable than EPS (especially when EPS is volatile)
 More useful for valuing companies with liquid assets (banks, etc.)
 Can be used when company is not a going concern
 Differences in P/B may relate to differences in long-run returns
Drawbacks of using P/B include:
 Some assets creating value (e.g. human capital) are not on the balance sheet
 Can be misleading when companies have different level of assets
 Accounting effects (e.g. R&D) can compromise BV
 BV is generally not reflective of market value
Price to sales (P/S): Ratio of market price per share to sales per share.
Rationale for using P/S include:
 Sales are less subject to accounting manipulation than EPS or BVPS
 Sales are positive even when EPS is negative
 Sales are more stable than EPS
 More useful in valuing mature, cyclical and zero income companies
 Differences in P/S may relate to differences in long-run returns
Drawbacks of using P/S include:
 A company may show sales growth despite low earnings and cash flow from
operations.
 P/S does not capture differences in cost structures.
 P/S does not capture different capital structures.
 Room for manipulating revenues remains.
Price to cash flow (P/CF): Ratio of market price to cash flow per share.
Rationales for using P/CF:
 Cash flow is less prone to manipulation than earnings.
 More stable than EPS
 Addresses the issue of accounting differences
 Differences in P/CF may relate to differences in long-run returns

Drawbacks of using P/CF

 When cash flow is defined as earnings plus noncash charges, items such as noncash
revenue and net changes in working capital are ignored.
 FCFE is viewed as more appropriate than cash flow. But, FCFE can be more volatile
or negative for certain businesses.
Dividend yield: Ratio of dividend rate to market price per share.

The formulas for trailing and leading dividend yield are given below:
Dividend rate
Trailing dividend yield =
Market price per share
where dividend rate = annualized amount of the most recent dividend
If the dividend is paid quarterly, then dividend rate = 4 * the most recent quarterly per-
share dividend.
Forecasted dividends per share over the next year
Leading dividend yield = Current market price per share
Rationales for using dividend yield:

 Dividend yield is a component of total return.
 Dividends are less risky than capital appreciation in total return.
Drawbacks of using dividend yield:
 Since dividend yield is only a part of total return, not using all the parts make this
valuation approach suboptimal.
 Dividend displacement of earnings is when investors choose to receive dividends
now over future earnings growth.
 The assumption that dividends are less risky than the capital appreciation
component implies that market prices are biased in the assessment of relative risk
of the two components.
Normalized EPS
Analysts address the issue of cyclicality in business by using normalized EPS. Normalized
EPS for cyclical companies are earnings expected under mid-cyclical conditions. Two main
methods for calculating normalized EPS are:
 Historical average method: Normalized EPS is calculated as the average EPS over
the most recent full cycle.
 Average ROE method: Normalized EPS = average full cycle ROE * current BV per
share
The ROE method is preferred because the historic average method does not account for
change in a business’s size.

Earnings yield (E/P)
Earnings yield (E/P) is the reciprocal of the P/E. A high E/P indicates an undervalued
security, whereas a low E/P indicates an overvalued security.
When stocks have zero or negative EPS, a ranking by earnings yield is meaningful, whereas
a ranking by P/E is not.
Justified multiples based on forecasted fundamentals
An analyst can compute ratios based on company fundamentals and compare with ratios
based on actual market price.
If the ratio based on actual market price is low then the stock is undervalued.
D1
P0 E 1−b
Forward P/E: = 1 =
E1 r − g r − g
P0 (1 − b)(1 + g)
Trailing P/E: =
E0 r−g
P0 ROE − g
Price to book value: =
B0 r−g
E0
P0 ( S0 ) (1 − b)( 1 + g)
Price to sales: =
S0 r−g
D0 r − g
Dividend yield: =
P0 1 + g
(1 + g)FCFE0
Value based on cash flow: V0 =
r−g
We can make the following inferences from the above formulas:
 P/E is directly related to expected earnings growth and inversely related to the
required rate of return.
 P/B is directly related to ROE. The larger the difference between ROE and r, the
higher the P/B.
 P/S is directly related to profit margin and growth, and inversely related to the
required rate of return.
 Dividend yield is directly related to the required rate of return and negatively
related to the expected rate of growth in dividends.
Predicted P/E based on cross-sectional regression
A predicted P/E can be estimated from a cross-sectional regression of P/E on the

fundamentals believed to drive security valuation. We can take several similar companies
and run a regression with the predicted P/E as the dependent variable and fundamental
characteristics such as the dividend payout ratio, beta, earnings growth rate, etc. as the
independent variables.
Example: Predicted P/E = 12.12 + (2.25 x DPR) – (0.20 X Beta) + (14.43 x EGR)
Predicted P/E is calculated by substituting the values of the variables in the estimated
regression equation. If the predicted P/E is less than the actual P/E, then the stock is
overvalued.
This method has three limitations:
 The model captures valuation relationships only for the specific stock over a
particular time period.
 Regression coefficients and explanatory power of regression tend to change
substantially over time.
 The method is prone to multicollinearity, or correlation within linear combinations
of the independent variables.
Valuation based on comparables
In the method of comparables, the analyst compares the price multiple of a stock with the
mean or media price multiple of comparable stocks. The method of comparables to
estimate the value of a company’s stock involves the following four steps:
1. Select and calculate the price multiple that will be used in the comparison.
2. Select the comparison asset or assets and calculate the value of the multiple for the
comparison assets. For a group of comparison asset, calculate the median or mean
value of the multiple for the assets. This multiple of the comparison asset(s) is called
the benchmark value of the multiple.
3. Adjust for differences in the fundamentals of stock and comparison stocks. Use the
benchmark value of the multiple to estimate the value of a company’s stock. Or,
compare the stock’s actual multiple with the benchmark value.
4. Fundamentals play an important role in the method of comparables as analysts can
use fundamentals to explain the differences between the estimated value of a
company’s stock and the current price of the company’s stock.
P/E – to – growth ratio (PEG)
PEG is calculated as the stock’s P/E divided by the expected earnings growth rate in
percent. Stocks with lower PEGs are more attractive than stocks with higher PEGs, all else
being equal. Some consider that a PEG ratio less than 1 is an indicator of an attractive value
level.

Some factors analysts must be aware of when using PEG are:

 PEG assumes a linear relationship between P/E and growth. But, the Gordon growth
model to derive a justified P/E shows the relationship is not linear.
 PEG does not account for differences in risk.
 PEG does not account for differences in the duration of growth. For instance, if the
denominator is the growth rate for a four-year period, it would not show differences
for longer term growth prospects.
Using P/Es to obtain terminal values in DDM
Terminal price multiple based on fundamentals: Divide both sides of the Gordon
growth model equation by the EPS. This gives the justified trailing terminal price multiple.
Terminal price multiple based on comparables: Determine the benchmark value, which
is the median industry P/E, the average industry P/E, or an average of its own historical
P/Es. The benchmark value is used to estimate the terminal P/E multiple using the formula
below:
P
Terminal value at time n: Vn = Benchmark value of trailing terminal E ∗ En
P
Terminal value at time n: Vn = Benchmark value of forward terminal E ∗ E(n+1)
Cash flows used in price and enterprise value multiples
The major cash flow measures used in calculating multiples are

 earnings plus noncash charges (CF),
 cash flow from operations (CFO),
 free cash flow to equity (FCFE), and
 earnings before interest, taxes, depreciation and amortization (EBITDA).
Important points to note:
 CF and EBITDA are not cash flow numbers because they do not account for noncash
revenue and net changes in working capital.
 In case of CFO, adjustments must be made for different accounting standards.
 FCFE is more closely linked to valuation theory. The disadvantage of FCFE is it is
more volatile than the CFO.
EV multiples
Enterprise value is measured as:

Enterprise value = Market value of common equity + Market value of preferred stock +
Market value of debt - Cash and investments
EV/EBITDA is one of the most widely used enterprise value multiples. It is the ratio of the

total company value to EBITDA. The rationales for using EV/EBITDA are as follows:
 For companies with different financial leverage, EV/EBITDA is a better choice than
P/E as it is a pre-interest figure.
 It increases the comparability across businesses as the age of the assets (and hence,
depreciation and amortization) may vary.
 EBITDA is often positive, whereas EPS can be negative.
The drawbacks for using EV/EBITDA are as follows:
 EBITDA overestimates cash flow from operations if working capital is growing.
 FCFF is more appropriate in valuation than EBITDA because it takes into account
capital expenditures. EBITDA will reflect capital expenditures only if depreciation
expenses match capital expenditures.
Other income measures used instead of EBITDA includes EBIT, EBITA, and FCFF.
EV/EBITDA is positively related to the expected growth rate in FCFF, profitability and
negatively related to the firm’s weighted average cost of capital.
Cross- border valuation comparisons
It is challenging to compare companies across countries because of differences in:

 accounting standards
 macroeconomic factors
 ability of companies to pass on the increase in costs to consumers
 cultural factors
 risk and growth opportunities
Momentum valuation indicators
Momentum indicators are valuation indicators that relate price to a fundamental such as
earnings to its own past values, or expected value. Three momentum indicators widely
used include:
 Earnings surprise: Earnings surprise (also called unexpected earnings) is the
difference between reported earnings and expected earnings.
 Standardized unexpected earnings (SUE): SUE is unexpected earnings divided by
the standard deviation in past unexpected earnings.
 Relative strength: Relative-strength indicators compare a stock’s performance
during a given period with its own past performance or with the performance of
some group of stocks.
Averaging P/E multiples
Using arithmetic means to calculate the P/E of a portfolio/index is inappropriate, because

the arithmetic mean is heavily influenced by outliers. The most appropriate measure to

calculate the average P/E is to use a weighted harmonic mean.

The expression for the weighted harmonic mean is:
1
XWH = w
∑i=1 i
N
Xi
where: wi = portfolio value weights summing to 1 and Xi > 0 for i = 1, 2, ……, n
Valuing a stock based on comparison of multiples
 If the price multiple of a stock is less than the benchmark value of the multiple, then the
stock is undervalued.
 If the price multiple of a stock is greater than the benchmark value of the multiple, then
the stock is overvalued.
 If the price multiple of a stock is equal to the benchmark value of the multiple, then the
stock is fairly valued.

LM05 Residual Income Valuation

Residual income, economic value added (EVA), and market value added (MVA)
Residual income is defined as the earnings for a given period minus the opportunity cost of
equity holders. It can be calculated in two ways:
Residual income = net income – (equity capital x cost of equity)
Residual income = EBIT (1 – tax rate) – (total capital x WACC)
EVA is a commercial implementation of the residual income concept.

EVA = NOPAT – (C% * TC)
where,
NOPAT = company’s net profit after taxes
C% = cost of capital
TC = total capital.
MVA is based on the idea that a company must generate economic profit for its market
value to increase.
MVA = Market value of the company – Accounting book value of total capital
Uses of residual income models
The residual income model is used to:

 value equity
 test goodwill impairment
 measure internal corporate performance
 determine executive compensation
Residual income model
According to the residual income model, the intrinsic value of equity is a sum of two
components:
 The current book value of equity
 The present value of expected future residual income
There are three ways to express the value of a stock under this model:
∞
RIt
V0 = B0 + ∑
(1 + r)t
t=1
∞
(Et − rBt−1 )
V0 = B0 + ∑
(1 + r)t
t=1

∞
(ROEt − r)Bt−1
V0 = B0 + ∑
(1 + r)t
t=1
As compared to the DDM and FCFE/FCFF models, residual income models are less sensitive
to terminal value estimates. This is because RI models include the company’s current book
value which usually represents a large portion of the estimated intrinsic value.
Fundamental determinants of residual income
The fundamental determinants of residual income are:

 book value of equity, and
 return on equity
This can be seen from the following expression:
RIt = (ROEt − r)Bt−1
Relationship between ROE and P/B
The justified price is the stock’s intrinsic value (P0 = V0 ).

ROE − r
P0 = B0 + ∗ B0
r−g
 If ROE > r, then residual income is positive, intrinsic value > book value, and justified
P/B > 1.
 If ROE = r, then residual income is zero, intrinsic value = book value, and justified P/B =
1.
 If ROE < r, then residual income is negative, intrinsic value < book value, and justified
P/B < 1.
Residual income valuation
In a multistage residual income model, we assume different growth rates for the short term
and the long-term. We forecast the residual incomes over a short term horizon and
estimate a terminal value based on the assumption made about continuing residual income
over the long term.
Continuing residual income is residual income after the forecast horizon. Usually, one of
the following assumptions is made for continuing residual income:
 Residual income continues indefinitely at a positive level.
 Residual income is zero from the terminal year forward.
 Residual income declines to zero as ROE reverts to the cost of equity through time.
 Residual income declines to some mean level.
If residual income continues indefinitely at a positive level, then we can use the following

formula (single stage model):

ROE − r
V0 = B0 + ∗ B0
r−g
If at the end of time horizon T, a certain premium over book value (PT – BT ) exists for the
company, then we can use the following formulas:
T
Et − rBt−1 PT − BT
V0 = B0 + ∑ +
(1 + r)t (1 + r)T
t=1
T
(ROE − r)Bt−1 PT − BT
V0 = B0 + ∑ +
(1 + r)t (1 + r)T
t=1
If residual income fades over time, we can use the following formula:
T−1
Et − rBt−1 ET − rBT−1
V0 = B0 + ∑ +
(1 + r)t (1 + r − ω)(1 + r)T−1
t=1
Where, ω = persistence factor between 0 and 1
1  residual income will not fade
0  residual income will not continue after the initial forecast horizon
The following exhibit reproduced from the curriculum indicates characteristics associated
with high and low levels of persistence:
Lower residual income persistence Higher residual income persistence
Extreme accounting rates of return (ROE) Low dividend payout
Extreme levels of special items (e.g., High historical persistence in the
nonrecurring items) industry
Extreme levels of accounting accruals
Implied growth rate
This equation can be used to calculate the implied growth rate in residual income given the
market P/B ratio and the required rate of return.
ROE − r
P0 = B0 + ∗ B0
r−g
Residual income models versus dividend discount and free cash flow models
The DDM and free cash flow to equity discount future cash flows, while the residual income
model starts with the book value of equity, and discounts the expected stream of residual
income available to equity shareholders.
As compared to the DDM and FCFE/FCFF models, residual income models are less sensitive

to terminal value estimates. This is because, RI models include the company’s current book
value which usually represents a large portion of the estimated intrinsic value.
Strengths and weaknesses of residual income models
Strengths of the residual income model include:

 The model gives less weight to terminal value.
 RI models use readily available accounting data.
 It can be used to value non-dividend paying companies.
 It can be used to value companies with no positive expected near-term free cash
flows.
 It can be used when cash flows are unpredictable.
Weaknesses of the residual income model include:
 The model is based on accounting data that is prone to manipulation.
 The accounting data may need adjustments.
 The model assumes that the clean surplus relation holds true.
 The model assumes that the cost of debt is equal to the interest expense.
Residual income models are most appropriate when:
 A company does not pay dividends.
 A company’s expected free cash flows are negative.
 When there is uncertainty in forecasting terminal values.
Residual income models are not appropriate when:
 The clean surplus relationship does not hold.
 The determinants of residual income such as book value and ROE are not
predictable.
Accounting considerations
When using residual income models, analysts must take the following into account:
1. Violations of the clean surplus relationship: Bt = B(t–1) + Et − Dt
2. Balance sheet adjustments for fair value
3. Intangible assets
4. Nonrecurring items
5. Other aggressive accounting practices
6. Differences in accounting standards across countries
Stock valuation using the residual income model
 If the intrinsic value based on residual income model is lower than the market price,
then the stock is overvalued.

 If the intrinsic value based on residual income model is equal to the market price, then
the stock is fairly valued.
 If the intrinsic value based on residual income model is greater than the market price,
then the stock is undervalued.

LM06 Private Company Valuation

Private versus public company valuation
Private companies differ from public companies in the following company-specific

factors:
 Stage in lifecycle – are generally in the early stages.
 Size – are generally smaller.
 Overlap of shareholders and management – have high managerial ownership.
 Quality/depth of management – have lower quality and less depth of management.
 Quality of financial and other information – disclose limited financial information.
 Pressure from short term investors – Less pressure, can focus on long term
decisions.
 Tax concerns – are motivated to report lower taxable income.
The effect of these factors on private company valuation is negative, except overlap of
shareholders and management, and pressure from short term investors.
Private companies differ from public companies in the following stock-specific factors:
 Liquidity – Shares are less liquid.
 Concentration of control – Control may be limited to one or two investors and
controlling shareholders may benefit at the cost of minority shareholders.
 Potential agreements restricting liquidity - Agreements may limit the ability to sell
shares.
The effect of all stock-specific factors on private company valuation is negative.
Reasons for private company valuation
The reasons for valuing private companies can be grouped into three categories:
Transaction-related valuations are required when selling or financing the firm.
Transaction types include:
 Private financing
 Initial public offering
 Acquisition
 Bankruptcy
 Share-based payment
Compliance-related valuations are required by law or regulation. Two primary focuses
are:
 Financial reporting
 Tax reporting

Litigation-related valuations are required to settle legal disputes such as:

 Shareholders disputes
 Damage claims
 Lost profit claims
 Divorce claims/ settlements
Financial analysts are mainly concerned with transaction related valuations for private
financing, IPO or acquisition.
Approaches to private company valuation
The three major private company valuation approaches are:

 Income approach: is based on the present value of expected future cash flows.
 Market approach: is based on the pricing multiples from the sale of similar assets.
 Asset approach: is based on the value of net assets.
Earnings normalization and cash flow estimation issues
Private company valuations may require adjustments to the reported earnings to estimate
normalized earnings of the company. Adjustments may be required for:
 non-recurring items
 discretionary expenses
 non-market level compensations
 personal expenses
In addition to earnings normalization, cash flow estimation is an important element of the
valuation process. Free cash flow (FCF) is the relevant concept of cash flow in this context.
Free cash flow to the firm (FCFF) represents free cash flow at the business enterprise level
and is used to value the firm. Equity value can be estimated based on firm value and debt.
Alternatively, free cash flow to equity (FCFE) can be used to value equity directly.
Income approach methods for valuation
Free cash flow method or the discounted cash flow method: The asset value is equal to
the present value of the estimate of the future cash flows and terminal value. The present
value of future cash flows is calculated using a risk-adjusted discount rate.
Capitalized cash flow method: Also known as the capitalized income method or
capitalization of earnings method. The value of an asset is equal to a single estimate of
economic benefits divided by an appropriate capitalization rate.
Excess earnings method:
Excess earnings = Normalized earnings – WACC x Tangible assets
The value of intangible assets is estimated by capitalizing excess earnings

Total value = value of tangible assets + value of intangible assets
Estimating discount rates
To estimate the discount rate we start with the CAPM model.

CAPM: Required return = risk-free return +βi (equity risk premium)
However, CAPM is suitable for large public companies. For small private companies,
additional premiums for small size and company-specific risk have to be added. These
adjustments are shown by the expanded CAPM model.
Expanded CAPM: Required return = Risk-free rate + Equity risk premium adjusted for beta
+ Small stock premium + Company-specific risk adjustment
Analysts can also use the build-up model to estimate the discount rate. Like the expanded
CAPM, risk premiums are added. However, no beta is used because it is assumed to be 1.
Analysts must include an industry risk premium to reflect the industry risk factor.
Build-up method: Required return on equity = Risk-free rate + Equity risk premium +
Small stock premium + Industry risk premium + Company-specific risk adjustment
If we are working with FCFF, then we need to discount these cash flows using the WACC. If
debt is available, the WACC will be based on both the cost of equity and the cost of debt. If
we determine that a small company will not be able to raise debt, then the WACC will be
equal to the cost of equity.
If we are evaluating an acquisition scenario, then the discount rate of the target should be
used to value the target. We should not use the discount rate of the acquirer.
Many private companies do not provide enough information to make accurate projections.
Therefore, the discount rate is adjusted upwards to reflect the projection risk.
Market approach methods
The three methods used in a market approach are:

Guideline public company method (GPCM): Value based on multiples of comparable
public companies; multiples are adjusted to reflect differences in the relative risk and
growth prospects.
The advantage is the availability of a large number of guideline companies with financial
and trading information. The disadvantage is that risk and growth adjustments to the
pricing multiple are subjective.
Guideline transactions method (GTM): Value based on pricing multiples derived from
the acquisition of control of entire public or private companies that were acquired.
Whereas GPCM uses a multiple that could be associated with trades of any size, GTM uses a
multiple that specifically relates to sales of entire companies.

The advantage is that it uses data from the acquisition of entire public companies instead of
small blocks of stock. The disadvantage is that data may be unreliable if the transactions
are not subject to public disclosure.
Prior transaction method (PTM) considers actual transactions in the stock of the subject
private company.
The advantage is it is most relevant since it is based on actual transactions on the
company’s stock. The disadvantage is that it can be less reliable if the transactions are
infrequent
Asset-based approaches
Underlying principle: Value = fair value of assets less the fair value of liabilities
The asset-based approach is rarely used for the valuation of going concerns because:
 limited market data is available to directly value intangible assets.
 it is difficult to value tangible assets such as special use plant and equipment.
Asset-based approach is appropriate for:
 companies which are winding up operations.
 holding companies such as REITS.
Valuation discounts and premiums
Two factors that affect the valuation of private companies are issues related to control and
marketability. When valuing the total equity of a private company, a control premium is
added if publicly traded companies are used as the basis for pricing multiple.
Discounts for lack of control are used to convert a controlling interest value into a non-
controlling equity interest value. The formula for a discount for lack of control is given by:
1
DLOC = 1 − [ ]
1 + Control premium
A discount for lack of marketability (DLOM) is an amount or percentage deducted from
the value of an ownership interest to reflect the relative absence of marketability.
DLOM can be estimated based on data from private sales of restricted stock in public
companies relative to their freely traded share price, stock sales prior to an IPO, and the
pricing of put options.
A DLOM may not be appropriate if there is a high likelihood of a liquidity event in the
immediate future.
DLOC and DLOM are multiplicative and not additive.
Total discount = 1 – [(1 − DLOC)(1 − DLOM)]

Fixed Income 2023 Level II High Yield Notes
LM01 The Term Structure and Interest Rate Dynamics

Spot rate and forward rate
Discount factor and spot rate: A spot rate is the interest rate on a security that makes a
single payment at a future point in time. The price or present value of a risk-free single-unit
payment (e.g., $1, €1, or £1) after N periods is called the discount factor, denoted by DFN.
The yield to maturity of the payment is the spot rate, denoted by ZN.
1
DF𝑁 =
[1 + Z𝑁 ]𝑁
Spot yield curve: The spot rate, ZN, for a range of maturities (for example: N = 1, 2, 3…) is
called the spot yield curve or spot curve. The spot curve shows, for various maturities, the
annualized return on an option-free and default-risk-free zero-coupon bond with a single
payment of principal at maturity.
Par curve: The par curve represents the yields to maturity on coupon-paying government
bonds, priced at par, over a range of maturities. It can be used to construct a zero-coupon
yield curve through a process called ‘bootstrapping’.
Forward rate and forward price: A forward rate is the interest rate that is determined
today for a loan that will be initiated at a future time period. fA,B-A represents a forward rate
of a loan initiated A years from today with tenor (further maturity) of B - A years. For
example, if A = 1 and B = 3, then f1,2 represents the forward rate on a two year zero issued 1
year from today.
FA,B-A represents the forward contract price for a zero-coupon bond with unit principal
which starts in the future at time A and ends at time B. For example, F1,2 represents the
forward contract price for a $1 par two-year zero issued one year from today.
Forward pricing and forward rate models
Forward pricing model: Forward contract price has to follow the equation below.
DF𝐵 = DF𝐴 × F𝐴,𝐵−𝐴
Forward rate model: The forward rate can be calculated using the following equation:
𝐵−𝐴
[1 + z𝐵 ]𝐵 = [1 + z𝐴 ]𝐴 [1 + f𝐴,𝐵−𝐴 ]
Two interpretations of forward rates
The forward rate model suggests two possible interpretations of forward rates. Suppose f1,2
= 12%, i.e., the rate agreed on today for a two-year loan to be made one year from today is
12%.
 First interpretation: 12% is the reinvestment rate that would make an investor

indifferent between buying a three-year zero-coupon bond versus investing in a

one-year zero-coupon bond and at maturity reinvesting the proceeds for two years.
In this sense, the forward rate can be viewed as a type of ‘breakeven interest rate’.
 Second interpretation: 12% is the two-year rate that can be locked in today by
buying a three-year zero-coupon bond rather than investing in a one-year zero-
coupon bond and, when it matures, reinvesting the proceeds in a zero-coupon
instrument that matures in two years. In this sense, the forward rate can be viewed
as ‘a rate that can be locked in’ by extending maturity by two years.
Relationships between spot and forward rates
 The spot rate for a given maturity can be expressed as a geometric average of the
short-term rate and a series of forward rates.
 If the spot curve is upward sloping the forward curve will lie above the spot curve. If
the spot curve is downward sloping the forward curve will lie below the spot curve.
If the spot curve is flat the forward rates will be equal to the spot rates.
 If the spot curve is upward sloping, the forward rate will rise as the initiation date,
A, for the forward contract is increased.
Yield to maturity (YTM)
 YTM is mathematically related to spot rates. It is the weighted average of spot rates
used in the valuation of bonds.
 The YTM is the discount rate that, when applied to a bond’s promised cash flows,
equates those cash flows to the bond’s market price.
 The YTM is the expected rate of return for a bond that is held until its maturity,
assuming that all coupon and principal payments are made in full when due and that
coupons are reinvested at the original YTM.
 The realized return is based on actual reinvestment rates and the yield curve at the end
of the holding period.
 The YTM can provide a poor estimate of expected return if:
o Interest rates are volatile.
o The yield curve is steeply sloped, either upward or downward.
o There is significant risk of default.
o The bond has one or more embedded options (e.g., put, call, or conversion).
Yield curve movement and forward curve
Forward prices and forward rates are based on current spot rates. The forward price
remains unchanged as long as future spot rates evolve as predicted by today’s forward
curve. However, if future spot rates do not follow the forward curve, the forward price will
change.
Active bond portfolio managers can outperform the bond market’s return by correctly

anticipating changes in interest rates relative to the projected evolution of spot rates.
Managers expecting lower future spot prices buy bonds in the forward market and vice
versa.
Riding the yield curve or rolling down the yield curve
 Active bond portfolio managers can outperform the bond market’s return by correctly
anticipating changes in interest rates relative to the projected evolution of spot rates.
 When a yield curve is upward sloping, the forward curve is always above the current
spot curve. If a trader does not believe that the yield curve will change its level and
shape over an investment horizon, he will buy bonds with a maturity longer than the
investment horizon. This strategy is called riding the yield curve or rolling down the
yield curve.
 If the view is correct, the trader’s total return will be greater than the return on a
maturity-matching strategy.
 The total return will depend on the spread between the forward rate and the spot rate
as well as the maturity of the bond. The longer the bond’s maturity, the more sensitive
its total return is to the spread.
The swap rate curve
The swap rate is the interest rate for the fixed-rate leg of an interest rate swap. The swap
rate curve (or swap curve) is the yield curve of swap rates. It provides another measure
for the time value of money.
The swap market is a highly liquid market because:
1. unlike bonds, a swap does not have multiple borrowers or lenders, only
counterparties who exchange cash flows and
2. swaps provide one of the most efficient ways to hedge interest rate risk.
Swap rates can be determined from the spot rates and spot rates can be determined from
the swap rates using the following equation:
T
s𝑇 1
∑ + =1
[1 + z𝑡 ]t [1 + z𝑇 )]T
t=1
Some of the reasons for the popularity of Libor/swap curves are:

 The swap spread reflects the default risk of private entities at a rating of about
A1/A+ (roughly the equivalent of most commercial banks).
 The swap market is unregulated. So swap rates are more comparable across
different countries relative to rates of government securities.
 The swap market has more maturities with which to construct a yield curve than do
government bond markets.

Spreads
The Swap spread is the spread paid by the fixed-rate payer of an interest rate swap over
the rate of the “on-the-run” (most recently issued) government security with the same
maturity as the swap.
The Z-spread is the constant basis point spread that would need to be added to the implied
spot yield curve so that the discounted cash flows of a bond are equal to its current market
price.
The TED spread is calculated as the difference between Libor and the yield on a T-bill of
matching maturity. The TED spread is an indicator of perceived credit risk in the general
economy.
The Libor-OIS spread is an indicator of the risk and liquidity of money market securities.
It is the difference between Libor and the overnight indexed swap (OIS) rate. An OIS is an
interest rate swap in which the periodic floating rate of the swap is equal to the geometric
average of an overnight rate (or overnight index rate) over every day of the payment
period. The index rate is typically the rate for overnight unsecured lending between
banks—for example, the federal funds rate for US dollars.
Traditional term structure theories
Unbiased expectations (Pure expectations theory):

 It states that the forward rate is an unbiased predictor of the future spot rate.
 The implication is that bonds of any maturities are perfect substitutes for each
other.
 For example, buying a bond with a maturity of five years and holding it for three
years has the same expected return as buying a three-year bond or buying a series
of three one-year bonds.
 The major limitation of this theory is that it simplistically assumes that investors are
risk neutral.
Local expectations:
 This theory is more rigorous than the unbiased expectations theory. Rather than
assuming that the bonds of any maturity will give the same returns, it assumes that
the expected return for every bond over the short run is the risk free rate.
 The limitation of the theory is that it is often observed that short-holding-period
returns on long-dated bonds do exceed those on short-dated bonds.
Liquidity preference theory:
 Liquidity preference theory states that liquidity premiums exist to compensate
investors for the added interest rate risk they face when lending long term and that

these premiums increase with maturity.

 Thus, the forward rate provides an estimate of the expected spot rate that is biased
upward by the amount of the liquidity premium.
 Existence of liquidity premiums implies that the yield curve will typically be upward
sloping.
 The limitation of this theory is that it fails to offer a complete explanation of the
term structure.
Segmented markets theory:
 This theory states that the yield of securities of a particular maturity is determined
entirely by the supply and demand for funds of that particular maturity.
 Each maturity sector is considered a segmented market, and the yield in each
segment is independent of the yield in other segments.
 The theory is consistent with a world where there are asset/liability management
constraints, either regulatory or self-imposed.
Preferred habitat theory:
 The preferred habitat theory is similar to the segmented markets theory in
suggesting that many participants have strong preferences for particular maturities.
 However, it does not state that yields at different maturities are determined
independently of each other.
 If the expected additional returns to be gained become large enough, institutions
will be willing to deviate from their preferred maturities or habitats.
A bond’s exposure to yield curve movement
 Shaping risk is defined as the sensitivity of a bond’s price to the changing shape of
the yield curve.
 Yield curve movements are well described by a combination of three independent
movements:
o Level
o Steepness
o Curvature
Studies have found that the first principal component (level) explained about 77%
of the curve movements, the second (steepness) explained 17% and third
(curvature) explained 3%.
 For active bond management, a bond investor may want to base trades on a
forecasted yield curve shape or may want to hedge the yield curve risk on a bond
portfolio.
 Shaping risk also affects the value of many options, which is very important because
many fixed-income instruments have embedded options.

Managing yield curve risks
Effective duration measures the sensitivity of a bond’s price to a small parallel shift in a
benchmark yield curve.
Key rate duration measures a bond’s sensitivity to a small change in a benchmark yield
curve at a specific maturity segment.
Consider a portfolio of 1-year, 5-year, and 10-year zero-coupon bonds with $100 value in
each position. The model for yield curve risk based on key rate durations can be expressed
as:
∆P
KeyDurFull = ( ) ≈ −KeyDur1 ∆z1 − KeyDur5 ∆z5 − KeyDur10 ∆z10
P
Calculating a measure based on parallel, steepness and curvature movements:
Let DL, DS, DC be the sensitivities of portfolio value to small changes in the level, steepness,
and curvature factors, respectively.
The proportional change in portfolio value that would result from a small change in the
level factor (ΔxL), the steepness factor (ΔxS), and the curvature factor (ΔxC) is given by the
following equation:
∆P
KeyDurFull = ( ) ≈ −KeyDurL ∆xL − KeyDurS ∆xS − KeyDurC ∆xC
P
The maturity structure of yield curve volatilities
Two reasons why it is important to quantify interest rate volatilities are:

1. Most fixed-income instruments and derivatives have embedded options. The value
of these instruments is impacted by changes in interest rate volatilities.
2. In order to manage the interest rate risk of fixed income securities it is important to
control the impact of interest rate volatilities.
The term structure of interest rate volatilities is a representation of the yield volatility of a
zero-coupon bond for different maturities. This volatility curve (“vol”) or volatility term
structure measures yield curve risk. Exhibit 12 from the curriculum illustrates the
historical volatility term structure for US treasuries.
Maturity
0.25 0.50 1 2 3 5 7 10 20 30
(years)
σ (t, T) 0.3515 0.3173 0.2964 0.2713 0.2577 0.2154 0.1885 0.1621 0.1332 0.1169
The term structure shows that typically short-term rates are more volatile than long-term
rates. Short-term volatility is strongly linked to uncertainty regarding monetary policy
whereas long-term volatility is strongly linked to uncertainty regarding the real economy
and inflation.

Developing interest rate views using macroeconomic variables
Bond risk premium is the expected excess return of a default-free long-term bond over an
equivalent short-term bond.
Many macroeconomic factors such as inflation, economic growth and monetary policy
influence bond pricing and required returns.
Monetary policy: During economic expansions, monetary authorities raise benchmark
rates to control inflation. This results in a bearish flattening; the short -term bond yields
rise more than the long-term bond yields. During economic recessions, monetary
authorities cut benchmark rates to stimulate economic activity. This results in a bullish
steepening; the short-term rates fall more than the long-term bond yields. Central banks
can also use asset purchases to influence the bond risk premium.
Other factors that influence the bond risk premium are:
 Fiscal policy: Greater budget deficits require more borrowing, which influences
both bond supply as well as the required yield. Yields rise when budget deficits rise
and vice versa.
 Maturity structure of debt: Longer government debt maturity structures are
associated with higher excess bond returns. The increased supply of long-term
maturity bonds raises the yield in that market segment.
 Investor demand: Greater demand from investors (such as pension funds and
insurance companies) increases prices and reduces the bond risk premium.

LM02 The Arbitrage-Free Valuation Framework

Arbitrage
An arbitrage opportunity refers to an opportunity to earn riskless profits without any net
investment of money. Arbitrage opportunities arise as a result of violations of the law of
one price. The law of one price states that two goods that are perfect substitutes must sell
for the same current price in the absence of transaction costs.
The two types of arbitrage opportunities are:
 Value additivity arbitrage opportunity: value of a portfolio does not equal the sum of
its parts.
 Dominance arbitrage opportunity: two risk free assets do not have the same return.
Arbitrage-free valuation is an approach to security valuation that determines security
values that are consistent with the absence of arbitrage opportunities.
 Using the arbitrage-free approach, any fixed-income security should be thought of
as a package or portfolio of zero-coupon bonds.
 Two bonds with the same maturity and different coupon rates are viewed as
different packages of zero-coupon bonds and valued accordingly.
 For bonds that are option free, an arbitrage-free value is simply the present value of
expected future values using the benchmark spot rates.
Binomial interest rate tree
An interest rate tree is a visual representation of the possible values of interest rates based
on an interest rate model and an assumption about interest rate volatility.
The possible interest rates must be consistent with the following three assumptions:
1. an interest rate model that governs the random process of interest rates
2. the assumed level of interest rate volatility
3. the current benchmark yield curve
Exhibit 5 from the curriculum presents a binomial interest rate tree.

Lognormal model of interest rates means:

 adjacent interest rates on the tree are multiples of e2σ
 interest rates cannot be negative
Two methods commonly used to estimate interest rate volatility:
 Historical method: volatility is calculated by using data from the recent past.
 Implied volatility: volatility is based on observed market prices of interest rate
derivatives, e.g. swaptions.
Pricing a bond using a binomial tree
To find the value of the bond at a particular node, we use the backward induction valuation
methodology.
Backward Induction: Start at maturity, fill in those values, and work back from right to
left to find the bond’s value at the desired node.
The value of the bond is given by the following equation:

C + (0.5 × VH + 0.5 × VL)
Bond value at a node =
1+i
Constructing the binomial interest rate tree
The binomial interest rate tree is fit to the current yield curve by choosing interest rates
that result in the benchmark bond value. We do this to ensure that the bond value is
arbitrage free. This process is called calibrating an interest rate tree to match a specific
term structure.
The steps are:
1. Determine forward rates.
2. Given a forward rate and a volatility assumption, determine the high and low rates.
To obtain low value multiply by e-σ. To obtain the high value multiply by e+σ.
3. Check if these rates result in the benchmark bond value.
4. If not, make adjustments to rates until you get the benchmark bond value.

Confirming the arbitrage-free value of a bond
To value an option free bond, we can either use the spot rates or a binomial tree. Since both
the methods are arbitrage free, the two values should be the same.
Consider an option-free bond with four years remaining to maturity, a coupon rate of 2%,
and a par value of $100. Assume spot rates are as shown in Exhibit 3.
Maturity (Years) One-Year Spot Rate
1 1.000%
2 1.201%
3 1.251%
4 1.404%
5 1.819%
Then the bond value can be calculated as:
$2 $2 $2 $100 + $2
1
+ 2
+ 3
+ = $102.33
(1.01) (1.01201) (1.01251) (1.01404)4
Next, consider a binomial interest rate tree calibrated to the same spot curve (Exhibit 13).
We can then use the backward induction discounting process to obtain the bond value.
Time 0 Time 1 Time 2 Time 3 Time 4
102.3254 102.6769 101.7639 101.1892 102
104.0204 102.8360 101.9027 102
103.6417 102.4380 102
102.8382 102
102
The tree produces the same value for the bond as the spot rates and is therefore consistent
with our standard valuation model.

Pathwise valuation
Pathwise valuation calculates the present value of a bond for each possible interest rate
path and takes the average of these values across paths.
The steps are:
1. Specify a list of all potential paths through the tree.
2. Determine the present value of a bond along each potential path.
3. Calculate the average across all possible paths.
Monte Carlo method
Monte Carlo methods are often used when a security’s cash flows are path dependent. The
Monte Carlo method involves randomly selecting paths in an effort to approximate the
results of a complete pathwise valuation.
The following steps are taken:
1. Simulate numerous paths of interest rates under some volatility assumption and
probability distribution.
2. Generate spot rates from the simulated future interest rates.
3. Determine the cash flow along each interest rate path.
4. Calculate the present value for each path.
5. Calculate the average present value across all interest rate paths.
To ensure that the model produces arbitrage free values a constant is added to all interest
rates on all paths such that the average present value for each benchmark bond equals its
market value. The constant added to all short interest rates is called a drift term. When this
technique is used, the model is said to be drift adjusted.
The Monte Carlo method is only as good as the valuation model used and the accuracy of
the inputs.
Mean reversion is often included in the simulation by imposing an upper and lower bound
on the random process for generating future interest rates.
Term structure models
Equilibrium term structure models

They describe the dynamics of the term structure using fundamental economic variables
that are assumed to affect interest rates. The characteristics of these models are:
 They can be one-factor or multifactor models.
 They make assumptions about the behavior of factors.
 They are, in general, more sparing with respect to the number of parameters that
must be estimated compared with arbitrage-free term structure models.

The two best known equilibrium models are:

CIR model: drt = k(Ɵ − rt )dt + σ√rt dZ
It assumes that interest rates are mean reverting and volatility increases as interest rates
increase.
Vasicek model: drt = k(Ɵ − rt )dt + σdZ
It has the same drift term as the CIR model. However, interest rates are calculated
assuming that volatility remains constant over the period of analysis. The main
disadvantage of the Vasicek model is that it is theoretically possible for the interest rate to
become negative.
Arbitrage-free models:
In arbitrage-free models, the analysis begins with the observed market prices of a
reference set of financial instruments and the underlying assumption is that the reference
set is correctly priced.
A benefit of arbitrage-free term structure models is that they are calibrated to the current
term structure. In contrast, equilibrium term structure models frequently generate term
structures that are inconsistent with current market data.
The Ho–Lee model: drt = θtdt + σdZ
The model is calibrated to market data by calculating the time-dependent drift term, θt,
from market prices. Because of the time dependency, there is a value for θt at each time
step, which is necessary for the model to produce prices that match market prices.
Kalotay–Williams–Fabozzi (KWF) Model: d ln(rt) = θtdt + σdZ
The KWF model is similar to the Ho-Lee model - it assumes constant drift, no mean
reversion, and constant volatility. However, we use the log of the short rate as the
dependent variable, which prevents interest rates from becoming negative.

LM03 Valuation and Analysis: Bonds with Embedded Options

Relationships between the values of a callable or putable bond, straight bond, and
embedded option
An embedded option represents a right that can be exercised by the issuer, by the
bondholder, or automatically depending on the course of interest rates. Embedded options
can be:
 Simple: call options, put options, etc.
 Complex: estate put, sinking fund bonds, etc.
A call option decreases the value of a bond to an investor. Therefore,
 Value of callable bond = Value of straight bond – Value of issuer call option
 Value of issuer call option = Value of straight bond – Value of callable bond
A put option increases the value of a bond to an investor. Therefore,
 Value of putable bond = Value of straight bond + Value of investor put option
 Value of investor put option = Value of putable bond – Value of straight bond
The value of any embedded option increases with interest rate volatility. The greater the
interest rate volatility, the higher the chances for the embedded option to be exercised.
A call option is more likely to be exercised when interest rates fall. Hence, the value of an
embedded call option is higher if the yield curve is downward sloping.
A put option is more likely to be exercised when interest rates rise. Hence, the value of an
embedded put option is higher if the yield curve is upward sloping.
Valuation of default-free callable and putable bonds in the absence of interest rate
volatility
In the absence of default and interest rate volatility, the bond’s future cash flows are
certain. Thus, the value of a callable or putable bond can be calculated by discounting the
bond’s future cash flows at the appropriate one-period forward rates, taking into account
the decision to exercise the option.
If a bond is callable, the decision to exercise the option is made by the issuer, who will
exercise the call option when the value of the bond’s future cash flows is higher than the
call price.
In contrast, if the bond is putable, the decision to exercise the option is made by the
bondholder, who will exercise the put option when the value of the bond’s future cash
flows is lower than the put price.
Valuation of default-free callable and putable bonds in the presence of interest rate

volatility
The procedure to value a bond with an embedded option in the presence of interest rate
volatility is as follows:
1. Generate a tree of interest rates based on the given yield curve and interest rate
volatility assumptions.
2. At each node of the tree, determine whether the embedded options will be
exercised.
3. Apply the backward induction valuation methodology to calculate the bond’s
present value. This methodology involves starting at maturity and working back
from right to left to find the bond’s present value.
Exhibits 11 and 12 from the curriculum illustrate the backward induction process used in
the valuation of the callable bond.
At nodes where the value of the bond exceeds the call price, we replace the value with the

call price.
Valuation of risky callable and putable bonds
There are two approaches for valuing risky bonds:

1. Value bonds by making the default probabilities explicit.
2. Value bonds by increasing the discount rates above the default-free rates to reflect
default risk.
To construct a suitable yield curve for a risky bond, we increase the one-year forward rates
derived from the default-free benchmark yield curve by a fixed spread which is estimated
from the market prices of suitable bonds of similar credit quality.
This fixed spread is called the zero-volatility or Z-spread. The Z-spread is appropriate to
value risky bonds with no embedded options when we assume that interest rates are static.
When valuing risky bonds with embedded options and when we assume interest rates are
volatile, the option-adjusted spread (OAS) is used. The OAS is the constant spread that,
when added to all the one-period forward rates on the interest rate tree, makes the
arbitrage-free value of the bond equal to its market price.
OAS
The option-adjusted spread is the single spread added uniformly to the one-period forward
rates on the tree to produce a value or price for a bond. This is depicted in Exhibit 14
below.
The OAS can be used to determine whether a bond is overvalued or undervalued relative to
another bond. Assume that Bond 1 and Bond 2 have similar characteristics.
 If the OAS of Bond 1 < OAS of Bond 2, then we can infer that Bond 1 is overpriced
(rich).
 If the OAS of Bond 1 > OAS of Bond 2, then we can infer that Bond 1 is underpriced
(cheap).
Exhibit 16 from the curriculum shows the effect of interest rate volatility on the OAS for a
callable bond.

OAS is sensitive to interest rate volatility:

 Higher the volatility, lower the OAS for a callable bond.
 If the volatility is 0, the OAS is the same as the Z-spread.
Duration
Effective duration indicates the sensitivity of a bond’s price to a 100 bps parallel shift of
the benchmark yield curve in particular, the government par curve, assuming no change in
the bond’s credit spread.
(PV− ) − (PV+ )
Effective duration =
2 x (ΔCurve) x (PVo )
The flowing procedure is used to apply this formula in practice.
1. Given a price (PV0), calculate the implied OAS to the benchmark yield curve at
appropriate interest rate volatility.
2. Shift the benchmark yield curve down, generate a new interest rate tree, and then
revalue the bond using the OAS calculated in Step 1. This value is PV–.
3. Shift the benchmark yield curve up by the same magnitude as in Step 2, generate a
new interest rate tree, and then revalue the bond using the OAS calculated in Step 1.
This value is PV+.
4. Calculate the bond’s effective duration.
Exhibit 20 from the curriculum compares the effective duration of option-free, callable and
putable bonds.

The effective durations of various types of instruments are shown in the table below.
Type of Bond Effective Duration
Cash 0
Zero-coupon bond ≈ Maturity
Fixed-rate bond < Maturity
Callable bond ≤ Duration of straight bond
Putable bond ≤ Duration of straight bond
Floater (Libor flat) ≈ Time (in years) to next reset
One-sided durations and key rate durations
One-sided durations: Because the prices of callable and putable bonds respond
asymmetrically to upward and downward interest rate changes of the same magnitude,
one-sided durations provide a better indication regarding the interest rate sensitivity of
bonds with embedded options than (two-sided) effective duration.
 One-sided up-duration measures the sensitivity of a bond’s value to an increase in
interest rates.
 One-sided down-duration measures the sensitivity of a bond’s value to a decrease in
interest rates.
Key rate durations are important to consider when dealing with non-parallel shifts in the
yield curve. Key rate durations reflect the sensitivity of a bond’s price to changes in specific
maturities on the benchmark yield curve.
Effective convexity
Since actual changes in bond prices are not linear it is useful to measure effective convexity,
i.e., the sensitivity of duration to changes in interest rates.
(PV− ) + (PV+ ) − [2 x (PVo ) ]
Effective convexity =
(ΔCurve)2 x (PVo )
Callable bonds have negative convexity at low interest rates (indicating limited upside) and

are similar to straight bonds at high rates.

Putable bonds are similar to straight bonds at low rates and have positive but low
convexity at high rates (indicating limited downside).
Valuation of a capped floater
A cap prevents the coupon rate from increasing above a specified maximum rate. A capped
floater protects the issuer against rising interest rates and is thus an issuer option.
Value of capped floater = Value of straight bond – Value of embedded cap
Exhibit 26 from the curriculum illustrates the valuation of the capped floater.
At each node, we check if the cap applies and if it does we adjust the cash flow accordingly.
Valuation of a floored floater
A floor prevents the coupon rate from decreasing below a specified minimum rate. A
floored floater protects the investor against declining interest rates and is thus an investor
option.
Value of floored floater = Value of straight bond + Value of embedded floor
Exhibit 28 from the curriculum illustrates the valuation of the floored security.

At each node, we check if the floor applies and if it does, we adjust the cash flows
accordingly.
Convertible bonds – defining features
Conversion price is the applicable share price at which the bondholders can convert their
bonds into common shares.
Conversion ratio reflects the number of shares of common stock that the bondholders
receive from converting their bonds into shares.
Conversion period is a predetermined period during which the conversion is allowed.
Components of a convertible bond’s value
There are a number of investment metrics and ratios that help analyze and value
convertible bonds.
The conversion value indicates the value of the bond if it is converted at the market price
of the shares.
The minimum value of a convertible bond is the greater of
 The conversion value and
 The value of the underlying option-free bond
The market conversion premium represents the price investors effectively pay for the
underlying shares if they buy the convertible bond and then convert it into shares. Scaled
by the market price of the shares, it represents the premium payable when buying the

convertible bond rather than the underlying common stock.

Market conversion premium per share = market conversion price – underlying share price
Market conversion premium ratio = Market conversion premium per share / Underlying
share price
Premium over straight value = (convertible bond price / straight value) – 1
Valuation of a Convertible Bond
Value of convertible bond = Value of straight bond + Value of call option on the issuer’s
stock
Value of callable convertible bond = Value of straight bond + Value of call option on the
issuer’s stock – Value of issuer call option
Value of callable putable convertible bond = Value of straight bond + Value of call option on
the issuer’s stock – Value of issuer call option + Value of investor put option
Comparison of the risk–return characteristics of a convertible bond, the straight

bond, and the underlying common stock
Exhibit 29 from the curriculum graphically depicts the price behavior of a convertible bond
and the underlying common stock.

LM04 Credit Analysis Models

Expected exposure, the loss given default, the probability of default, and the credit
valuation adjustment
The three major factors considered while modelling credit risk are:
1. Expected exposure to default loss: This is the maximum amount of money that an
investor could lose in the event of a default before factoring in recovery.
2. Recovery rate: This is the percentage of the loss recovered from a bond in the event
of a default. From the recovery rate, we can also calculate the loss severity or the
loss given default; loss given default = 1 – recovery rate.
3. Probability of default: The probability that a bond issuer will not meet its
contractual obligations.
When calculating expected loss:
 risk-neutral probability should be used (this is greater than the actual default
probability).
 discounting should happen at the risk-free rate.
Credit valuation adjustment: The CVA is the value of credit risk in present value terms. It
represents the compensation for bearing credit risk.
Credit scores and credit ratings
Credit scores and credit ratings are third party evaluations of credit worthiness.
Credit scores:
 Credit scores are used in retail lending and they are focused on small businesses and
individuals.
 Credit scoring methods vary across countries.
 FICO score is used in the US. The scores range from 300 to 850 (an ordinal rating
focused on the probability of default). It is calculated based on five primary factors:
1. Payment history
2. Debt burden
3. Length of credit history
4. Types of credit used
5. Recent searches for credit
Credit ratings:
 Credit ratings are used in the wholesale markets for bonds issued by corporations
and government entities.
 The three major global credit rating agencies are Moody’s Investors Service,
Standard & Poor’s, and Fitch Ratings.

 Like credit scores, credit ratings are also on an ordinal scale and focus on the
probability of default.
 Each agency provides credit ratings for issuers as well as specific issues. The issuer
rating is typically for senior unsecured debt. Specific issues can have different
ratings. The ratings are notched up or down to reflect the priority of claims for
specific debt issues. Notching provides an indication of expected LGD.
Expected return on a bond given transition in its credit rating
We can use credit ratings and a transition matrix of probabilities to adjust a bond’s YTM to
reflect the probabilities of credit migration.
 For each possible transition, we can calculate the expected price change as the
product of modified duration and the change in the spread.
 The expected percentage change in bond value is found by multiplying each
expected percentage price change by its respective transition probability and
summing the products.
Credit spread migration typically reduces the expected return for two reasons:
1. Probabilities for change are skewed towards a downgrade.
2. Change in credit spread is much larger for downgrades then for upgrades.
Structural versus reduced-form models
Structural Models Reduced-Form Models
What is it? Predict why a default may occur. Predict when a default may occur
Based on insights from option (default time).
pricing theory. Statistical methods are used.
The values for debt and equity at Default intensity (probability of
time T can be expressed as: default over the next period) is
estimated using regression analysis
D(T) + E(T) = A(T)
E(T) = Max[A(T) – K,0] on company-specific and macro-
D(T) = A(T) – Max[A(T) – K,0] economic variables.
Default is an exogenous variable that
Probability of default is
endogenous to the model. occurs randomly.
Assumptions Assets are actively traded. Inputs are observable variables
Liabilities easily determined. including historical data.
Strengths Economic explanation for default. Practical assumptions.
Business cycle is considered.

Weaknesses Unrealistic assumptions. Economic reasons for default are not

Business cycle not considered. explained.
Default is a surprise and can happen
at any time.
Used by Company managers, commercial Financial analysts trying to value debt
bankers, rating agencies. (I.e., securities. (The information required
entities that have access to the is easily available to external parties.)
internal information needed for
this model.)
Value of a bond and its credit spread, given assumptions about the credit risk
parameters
For the credit analysis of a risky bond in a volatile interest rate environment, we use the
arbitrage-free framework.
The first step in the arbitrage-free framework is to build the binomial interest rate tree
under assumption of no arbitrage. Once the tree is built we need to verify that it is correctly
calibrated.
Analyzing a fixed-coupon corporate bond:
This tree can then be used to analyze a fixed-coupon corporate bond. The steps are:
1. Determine value of bond assuming no default (VND)
2. Calculate credit valuation adjustment (CVA)
3. Fair value of bond = VND – CVA
4. Using fair value determine YTM.
5. Using YTM determine credit spread.
Analyzing a floater:
We can also use the arbitrage-free framework to analyze a floater. The process is similar to
the analyzing a fixed coupon security. The steps are:
1. Calculate the VND given the quoted margin (QM).
2. Calculate the CVA.
3. Fair value = VND – CVA
4. Use trial and error to determine the discount margin (DM).
Impact of change in interest rate volatility: A change in the assumed level of interest
rate volatility has a small impact on the fair value of a corporate bond.
Interpreting changes in credit spreads
Credit spreads capture microeconomic factors that concern the issuer and the specific issue
itself. They include:

 Expected loss due to default. (This is the largest component.)

 Liquidity and tax differences between corporate and benchmark government bonds.
 Compensation for uncertainty regarding expected loss due to default.
The arbitrage-free framework and credit risk model can be used to study the relationship
between default probability, recovery rate and credit spread. If we tweak these models and
use the assumed binomial interest rate tree, assumed recovery rate and ‘observed credit
spreads’ as inputs, we can get the ‘implied POD’ as an output. This implied POD will be the
risk-neutral default probability.
Risk-neutral probability will be higher than actual default probability because it includes
the following:
1. Credit migration risk
2. Uncertainty over timing of possible default loss
3. Liquidity and tax considerations
Credit spreads are more sensitive to changes in the hazard rate (POD) than to changes in
the recovery rate.
Rating agencies use ‘notching’ to address differences in recovery rate between bonds.
The term structure of credit spreads
The main determinants of the term structure of credit spreads are:

 Credit quality
 Financial conditions
 Market supply and demand dynamics
 Issuer or industry-specific factors
The shape of the credit curve term structure is heavily influenced by market expectations
of defaults over time.
 Flat: If the market expectation is that the POD will stay the same over time, then we
get a flat curve.
 Upward sloping: If the market expectation is that the POD will increase over time,
then we get an upward sloping curve.
 Downward sloping: If the market expectation is that the POD will decrease over
time, then we get a downward sloping curve.
Based on empirical evidence:
 For investment grade bonds, the credit curve is typically slightly upward sloping.
 For high-yield issuers, we could either have an upward sloping or a downward
sloping credit curve.
 For distressed securities that are likely to default the credit curve is not a reflection

of credit risk across maturities.
Credit analysis of securitized debt
Credit analysis of structured debt requires a different approach compared to credit analysis
of other risky bonds. There are three major factors to consider when evaluating asset-
backed securities (ABS):
1. Underlying collateral -
 Granularity and homogeneity describe the underlying collateral. Granularity refers
to the number of obligations in the overall structured financial instrument.
Homogeneity refers to the degree to which the underlying debt characteristics
within a structured financial instrument are similar across individual obligations.
 Three major credit analysis approaches can be used for ABS: book of loans,
portfolio, loan by loan. The appropriate approach depends on the asset type, tenor
and granularity/homogeneity of the underlying.
2. Origination and servicing of collateral - Two major questions asked here are:
 How good is the servicer at managing and servicing the portfolio over the life of the
transaction?
 What is the track record of the servicer?
3. Structure of the transaction - Two major questions asked here are:
 What is the relationship between the issuer (SPE) and the originator?
 What credit enhancements are in place?
Covered bonds: Covered bonds are senior debt obligations issued by a financial
institution. The unique feature of covered bonds is that they give recourse to the
originator/issuer in addition to the predetermined underlying collateral.
When evaluating covered bonds, we should consider both the dual recourse principle as
well as the quality of the underlying asset pool.

LM05 Credit Default Swaps

CDS description; single name and index CDS
A credit default swap is a derivative contract between two parties, a credit protection
buyer and credit protection seller, in which the buyer makes a series of cash payments to
the seller and receives a promise of compensation for credit losses resulting from the
default of a third party.
The exhibit below reproduced from the curriculum shows the structure of payment flows.
The ISDA master agreement is a document that lays down the rules that all CDS contracts
must conform to and other guidelines for the functioning of the CDS market.
Each CDS contract has a notional amount and maturity date.
The debt of the third party on which the CDS is written, is called the reference entity. The
CDS covers all issues of the reference entity with similar or higher seniority. The
underlying is the credit quality of a borrower.
The payoff on the CDS is determined by the cheapest-to-deliver obligation, which is a debt
instrument with the lowest cost but same seniority as the reference obligation.
There are three types of CDS:
 Single-name CDS is associated with one specific borrower.
 Index CDS is associated with an equally weighted combination of borrowers.
o Higher the credit correlation between borrowers, higher the cost of protection.
o When an entity within an index defaults, that entity is removed from the index
and settled as a single- name CDS based on its relative proportion in the index.
 Tranche CDS covers borrowers only up to certain levels of losses.
Credit events and settlement protocols
Credit event is an event that defines default by the reference entity. When a credit event
occurs, the protection seller makes a payment to the protection buyer.

Three general types of credit events include:

 Bankruptcy
 Failure to pay
 Restructuring
If credit event has occurred, two parties to a CDS have the right to settle the CDS.
Settlement can happen in two ways:
 Physical settlement: The debt instrument (reference obligation) is delivered by the
protection buyer to the protection seller in exchange for a payment equal to the
notional amount of the CDS contract.
 Cash settlement: The credit protection seller pays cash to the credit protection
buyer as determined by the cheapest-to-deliver obligation of the reference entity.
Default does not mean that the creditor will lose the entire amount owed as a portion of the
loss could be recovered. The recovery rate is the percentage of the loss recovered. The
payout ratio is an estimate of the expected credit loss.
Payout ratio = 1 – recovery rate (%)
Payout amount = Payout ratio * notional amount
Principles and factors which influence CDS pricing
Basic pricing concepts:

 Probability of default: probability of non-payment of an upcoming interest or
principal obligation. Since CDS typically cover a multi-year horizon, we use the
hazard rate to calculate the probability of default for each year.
 Hazard rate: probability that an event will occur given that it has not already
occurred. It is a conditional probability. Ex: death.
 Probability of survival: Given the hazard rate, the probability of survival is
calculated as: 1 – hazard rate.
 Loss given default: amount that will be lost if a default occurs.
 Expected loss: full amount owed minus the expected recovery. It is also given by this
formula:
Expected loss = loss given default ∗ probability of default
Example: Assume that the hazard rate for a 10 year bond is 2% each year. The probability
of no default in 10 years is (0.98) × (0.98) … (0.98) = (0.98)10 = 0.817. Thus, the probability
of default is 1 – 0.817 = 0.183, or 18.3%.
Pricing a CDS means determining the CDS spread or upfront payment given a particular
coupon rate for a contract.
The upfront payment of a CDS is calculated as the difference in the present value of the

protection leg and the present value of the premium leg.

Upfront payment = present value of protection leg − present value of premium leg
Protection leg is the payment from the credit protection seller to the buyer when a default
occurs. Premium leg is a series of payments from the protection buyer to the protection
seller until default occurs. The party with the greater present value makes a payment to the
counterparty.
CDS prices are often quoted as credit spreads. The convention in the CDS market for the
fixed payments made from the CDS buyer to the CDS seller is to use standardized coupons
of 1% for investment-grade debt and 5% for high-yield debt. If the credit spread of the
reference obligation is different from these rates, then an upfront payment is made from
one party to the other.
Credit spread ≈ (Upfront premium/Duration) + Fixed coupon
PV of credit spread = Upfront premium + PV of fixed coupon
Upfront premium ≈ (Credit spread – Fixed coupon) x Duration
Price of CDS in currency per 100 par = 100 − Upfront premium%
The change in the value of CDS after inception can be calculated as:
Profit for the buyer of protection ≈ Change in spread in bps ∗ Duration ∗ Notional Amount
Use of CDS to manage credit exposure
The main objective of using a CDS is to increase or decrease credit exposure.

 Lender buys CDS to decrease credit exposure to a borrower.
 Dealer sells CDS to diversify or to hedge another exposure.
The credit curve represents credit spreads for a range of maturities of a company’s debt. It
is determined by CDS rates and is affected by a number of factors including the hazard rate.
 Flat credit curve implies a constant hazard rate.
 Upward slowing curve implies a greater likelihood of default in later years.
 Downward-sloping credit curve implies a greater probability of default in earlier
years.
CDS trading strategies include:
 Curve trade: This involves buying and selling protection on the same reference
entity but with different maturities. For example, if the credit curve is expected to
flatten, then an investor should buy protection (go short) in the short-term, and sell
protection (go long) in the long-term.
 Long/short trade: This involves buying protection for one reference entity and
selling protection for another reference entity. The reference entities may be from
the same industry or related to each other in some other way. For example – Honda

and Toyota.
Use of CDS to exploit valuation discrepancies
Valuation differences in CDS arise because of different opinions of the price of credit risk. It
is possible that the credit spread implied by the bond market is different from the credit
spread implied by the CDS market. A difference in the credit spread in these two markets is
the foundation of a strategy known as basis trade.
By exploiting any temporary mispricing between the two markets, an investor can capture
a gain as the spreads eventually converge.
Suppose the bond market implies a 5% credit risk premium whereas the CDS market
implies a 4% credit risk premium.
To profit from this situation, an investor can:
 Buy the CDS (purchase credit protection)
 Buy the bond
This concept can be extended to take advantage of valuation disparities in several
instruments such as equities, synthetic CDO etc.
Suppose Company A will issue significant new debt (which will increase its probability of
default) and used the funds to acquire Company B shares at a premium.
To profit from this situation, an investor can:
 Buy protection on Company A debt
 Buy Company B shares

Derivatives 2023 Level II High Yield Notes
LM01 Pricing and Valuation of Forward Commitments

Forward contracts and arbitrage
The arbitrage free forward price is given by:
F0 (T) = S0 (1 + r)T
Example: An asset has a current spot price of 100. There are no cash flows associated with
this asset. You enter into a contract to sell this asset at the end of one year. What is the
arbitrage-free forward price assuming a risk-free rate of 5%?
F0 (1) = 100(1 + 0.05)1 = 105
If a forward contract is overpriced, we can use carry arbitrage to generate a riskless
profit. The steps are:
Step 1. Sell the forward contract on the underlying.
Step 2. Purchase the underlying.
Step 3. Borrow the funds for the underlying purchase.
Step 4. Borrow the arbitrage profit.
If a forward contract is underpriced, we can use reverse carry arbitrage to generate a
riskless profit. The steps are:
Step 1. Buy the forward contract on the underlying.
Step 2. Sell the underlying short.
Step 3. Lend the short sale proceeds.
Step 4. Borrow the arbitrage profit.
Instructor’s Note:
We always ‘buy low and sell high’. If the forward contract is overpriced, we will sell the
forward contract and purchase the underlying. However, if the forward contract is
underpriced, we will sell the underlying and purchase the forward contract.
The value of the forward contract at time t is given by:
Vt = Present value of difference in forward price = PV [Ft(T) – F0(T)]
Example: Assume that we entered into a one-year forward contract with price F0 (T) = 100.
Nine months later (t = 0.75), the observed price of the underlying is 105 and the interest
rate is 5%. What is the value of the existing forward contract expiring in three months?
Solution:

F0 (T) = 100, S0.75 = 105, r = 5%, T − t = 0.25

The three-month forward price at t: Ft (T) = 105 (1 + 0.05)0.25 = 106.2886. The value of
106.2886 − 100
the existing forward contract is: Vt = (1+0.05)0.25
= 6.2123.
Carry arbitrage model when underling has cash flows

Let 𝛾 denote carry benefits such as dividends or bond coupon payments arising from the
underlying instrument during the life of the contract. θ denotes carry costs such as storage
costs during the life of the contract.
The forward price can now be expressed as:

F0(T) = FV(S0 + θ0 – γ0) or,
F0(T) = S(1 + r)T + FV of costs – FV of benefits
Example: Consider a stock which is selling for $100. The stock will pay a dividend of 2.9277
at t = 0.5. The risk-free rate is 5%. What is the arbitrage-free price of a one-year forward
contract on this stock?
Solution: Using the first formula,

F0(T) = FV(S0 + θ0 – γ0)
F0(T) = FV(100 + 0 – 2.9277/1.050.5) = (100 + 0 – 2.8571) x 1.051 = 102
Using the second formula,
F0(T) = S(1 + r)T + FV of costs – FV of benefits
F0(T) = 100 x 1.051 + 0 - 2.9277 x 1.050.5 = 105 + 0 – 3 = 102
The forward price with continuous compounding is:
F0 (T) = S0 e(rc +θ−γ)T
The value of the forward contract at time t is given by:
Vt = Present value of difference in forward price = PV [Ft(T) – F0(T)]

FRA pricing and valuation

A forward rate agreement is an over-the-counter forward contract in which the
underlying is an interest rate. FRAs are usually expressed in the “X x Y” convention, ‘X’
represents the point where the underlying loan starts. This also the point where the FRA
expires. ‘Y’ represents the point where the underlying loan ends. A 3 x 9 FRA is depicted in
the figure below.
The FRA fixed rate, i.e., price of an FRA can be calculated using these steps:
1. Set up the time line.
n2
1+(r2× )
360
2. Compute the de-annualized fixed rate: n1 −1
1+(r1× )
360
3. Annualize by multiplying by 360/(n2 - n1)
The FRA value for pay fixed receive floating can be calculated using these steps:
1. Find the new FRA fixed rate at time t
2. Compute payoff at n2: (FRAt – FRA0) x (n2 – n1)/360 x NP
3. Discount back to time t.
Example: Suppose we entered a receive-floating 6×9 FRA at a rate of 0.89%, with a notional
amount of $5,000,000 at T = 0. After 90 days, the three-month US dollar Libor is 1.10% and
the six-month US dollar Libor is 1.25% which will be the discount rate to determine the
value. What is the value of the original receive-floating 6×9 FRA?
Solution:
1. We first compute the new FRA rate at time t.
Step 1: Set up the time line.

Step 2: Compute the de-annualized fixed rate:

180
1 + (r2 × n2/360) 1 + (0.0125 × )
−1= 360 − 1 = 0.00349
1 + (r1 × n1/360) 90
1 + (0.0110 × 360)
Step 3: Annualize by multiplying by 360/(n2 - n1)

360
0.00349 × = 0.01396 = 1.396%
180 − 90
2. Compute payoff at n2: (FRAt – FRA0) x (n2 – n1)/360 x NP
90
(0.01396 − 0.0089) × × $5million = $6,325
360
3. Discount back to time t.
$6325
= $6,285.71
1 + 0.0125 × 180/360
Fixed-income forward and futures contracts

Unique issues:
 Quoted priced, accrued interest and full price: The quoted price is the clean price
without the interest accrued since the last coupon date. The full price is the quoted
price plus interest accrued since the last coupon date.
 Conversion factor: Fixed-income futures contracts often have more than one bond
that can be delivered by the seller. Because bonds trade at different prices based on
their maturity and stated coupon, an adjustment known as the conversion factor is
used to make all deliverable bonds roughly equal in price.
 Cheapest to deliver bonds: However, even after applying the conversion factor there
will be a bond that is the cheapest to deliver.
The fixed-income forward or futures price can be expressed as:
F0 (T) = FV0,T (B0 + AI0 ) − AIT − FVCI0,T
The quoted price which includes the conversion factor is:
QF0 = F0 (T)/CF
Example: Suppose T = 0.25, CF = 0.8, B0 = 107 (quoted price), FVCI0,T = 0.0 (meaning no
accrued interest over the life of the contract), r = 0.2%, AI0 = 0.07 and AIT = 0.20.

The futures price is:

F0 (T) = FV0,T (B0 + AI0 ) − AIT − FVCI0,T
F0 (T) = (107 + 0.07)(1 + 0.002)0.25 − 0.20 − 0.0 = 106.9235
The quoted futures price is:
QF0 = F0 (T)/CF
106.9235
QF0 = = 133.6544.
0.8
Interest Rate Swaps
A swap is an over-the-counter contract between two parties to exchange a series of cash
flows based on some pre-determined formula. In a plain vanilla interest rate swap, one
party pays fixed rate, whereas the other party pays a floating rate.
Swap fixed rate = (1 – final discount factor) / (sum of all discount factors)
Value of a fixed rate swap at Time t = Sum of the present value of the difference in fixed
swap rates x Stated notional amount
V = NA(FS0 − FSt )(d1 + d2 + d3 + ⋯ + dn )
Example: Suppose two years ago we entered a €100,000,000 seven-year receive-fixed

Libor-based interest rate swap with annual resets using 30/360 day count. The fixed rate
in the swap contract at initiation was 3%. Using the present value factors in the below
given table, what is the value for the party receiving the fixed rate?
Maturity (years) Present Value Factors
1 0.9906
2 0.9789
3 0.9674
4 0.9556
5 0.9236
Solution: We first compute the new swap fixed rate. The sum of the present values is
4.8161. Therefore, the fixed swap rate is:
1−0.9236
rFIX = 4.8161
= 1.59%.
V = NA(FS0 − FSt )(d1 + d2 + d3 + ⋯ + dn )

V = (0.03 − 0.0159)4.8161 × 100,000,000 = €6,790,701.
Thus, the swap value is €6,790,701.
Currency Swaps
In a currency swap, two counterparties agree to exchange future interest payments in

different currencies. These interest payments can be based on either a fixed interest rate or
a floating interest rate. Generally, there is no netting.
Pricing a currency swap involves solving for the appropriate notional amount in one
currency, given the notional amount in the other currency, as well as two fixed interest
rates such that the currency swap value is zero at initiation.
The value of a currency swap at time t is given as:
Va = FBa − St FBb
The value of fixed rate bonds is computed as:
FBk = NAk (rFIX,k (d1 + d2 + d3 + ⋯ dn ) + dn )
Example: A US company needs to borrow 100 million Australian dollars (A$) for one year
for its Australian subsidiary. The company decides to issue US-denominated bonds in an
amount equivalent to A$ 100 million. The company enters into a one-year currency swap
with quarterly reset using 30/360 day count and exchanges the notional amounts at
initiation and at maturity. Based on the below interbank rates and A$/US$ spot exchange
rate of 1.160:
Days to Maturity A$ interest rates US$ interest rates
90 2.35% 0.25%
180 2.45% 0.30%
270 2.60% 0.35%
360 2.75% 0.40%
1) What is the fixed swap rate for Australian and US dollars?
2) What is the notional amount in US dollars?
3) What are the fixed swap annual payments in the currency swap?
Solution to 1:
Days to A$ Calculation Present US$ Calculation Present
Maturity interest Value interest Value
rates A$ rates US$
90 2.35% 1 0.9942 0.25% 1 0.9994
90 90
1 + 0.0235 (360) 1 + 0.0025 (360)
180 2.45% 1 0.9879 0.30% 1 0.9985

180 180
1 + 0.0245 (360) 1 + 0.0030 (360)
270 2.60% 1 0.9809 0.35% 1 0.9974

270 270
1 + 0.0260 (360) 1 + 0.0035 (360)

360 2.75% 1 0.9732 0.40% 1 0.9960

360 360
1 + 0.0275 (360) 1 + 0.0040 (360)
The Australian dollar periodic rate is:

1 − 0.9732
= 0.6809%
0.9942 + 0.9879 + 0.9809 + 0.9732
The annualized rate is 0.6809% × 360/90 = 2.7236%.
The US dollar periodic rate is:
1 − 0.9960
= 0.1002%
0.9994 + 0.9985 + 0.9974 + 0.9960
The annualized rate is 0.1002% × 360/90 = 0.4009%.
Solution 2:
The US dollar notional amount is: A$ 100,000,000 / 1.16 = US$ 86,206,897.
Solution 3:
Based on the fixed rate calculated in Solution 1, the fixed swap payments for A$ is:
A$ 100,000,000 × (90/360)(0.027236) = A$ 680,900.
Based on the fixed rate calculated in Solution 1, the fixed swap payments for US$ is:
US$ 86,206,897 × (90/360)(0.004009) = US$ 86,400.86.
Example: This example builds on the previous example addressing currency swap pricing.
All the other details remain the same but now 60 days have passed since the initiation of
the currency swap and we observe the following market information:
Days to maturity Present value A$ Present value US$
30 0.9954 0.9995
120 0.9932 0.9983
210 0.9876 0.9967
300 0.9841 0.9951
Sum 3.9603 3.9896
What is the current value of the currency swap entered into 60 days ago if the current spot
exchange rate is A$/US$ = 1.15?
Solution: Based on the data given, the currency swap value is:
Va = FBa − St FBb
Va = 100,000,000(0.006809 (3.9603) + 0.9841) – 1.15(86,206,897)(0.001002(3.9896) +
0.9951) = A$ 2,058,100.85
Equity Swaps
An equity swap is an over-the-counter derivative contract in which two parties agree to

exchange a series of cash flows whereby one party pays a variable series that will be
determined by an equity index/security and the other party pays either a variable series
determined by a different equity or a fixed series.
The cash flows for the three types of equity swaps can be expressed as follows:
1. Receive-equity, pay-fixed: Notional Amount x (Equity return – Fixed rate)
2. Receive-equity, pay-floating: Notional Amount x (Equity return – Floating rate)
3. Receive-equity (a), pay-equity (b): Notional Amount x (Equity return (a) – Equity
return (b)) where a and b denote different equities.
Equity Swap Pricing and Valuation

The equity swap value for a receive-fixed, pay-equity is determined as follows:
Vt = FBt(C0) – (St/St–)NAE – PV(Par – NAE)
Example: Suppose six months ago we entered a receive-fixed, pay-equity five-year annual
reset swap in which the fixed leg is based on a 30/360 day count. The swap was entered at
a fixed rate of 2.0%, the notional was 5,000,000 and the equity was trading at 100. The
current spot rates have fallen to 1.5% across maturities and the equity is trading at 108.
What is the fair value of the equity swap?
Solution: The value of the equity swap is expressed as:
Vt = FBt(C0) – (St/St–)NAE – PV(Par – NAE)
Since the par value of the bond is equal to the notional amount of equity, the third term in
the expression reduces to 0.
The value of the implied fixed-rate bond FBt(C0) is computed as:
Years Present value Factor Fixed Cash Flow PV
0.5 1 5,000,000*0.02 = 100,000 99,260

= 0.9926
1 + 0.015 × 0.5
1.5 1 5,000,000*0.02 = 100,000 97,790
= 0.9779
1 + 0.015 × 1.5
2.5 1 5,000,000*0.02 = 100,000 96,390
= 0.9639
1 + 0.015 × 2.5
3.5 1 5,000,000*0.02 = 100,000 95,010
= 0.9501
1 + 0.015 × 3.5
4.5 1 5,000,000 + (5,000,000*0.02) = 4,777,680
= 0.9368
1 + 0.015 × 4.5 5,100,000
The total of the present value of the fixed cash flows i.e. the value of the bond is 5,166,130.
Vt = FBt(C0) – (St/St–)NAE
Vt = 5,166,130 - (108/100) x 5,000,000 = -233,870.

LM02 Valuation of Contingent Claims

Binomial Model: No Arbitrage Approach (Call Options)
The one-period binomial model can be understood using the figure below:
The no-arbitrage single-period valuation equation for call options is:

c = hS + PV(–hS– + c–) or equivalently,
c = hS + PV(–hS+ + c+)
where the hedge ratio h is given by the expression:
c+ − c−
h= +
S − S−
Based on these expressions, we can say that a call option is equivalent to owning h shares
of stock partially financed, where the financed amount is PV(–hS– + c–).
Example: A non-dividend-paying stock is currently trading at $50.00. A call option has one
year to mature, the periodically compounded risk-free interest rate is 7%, and the exercise
price is $50.00. Assume a single-period binomial option valuation model, where u = 1.25
and d = 0.80. Estimate the option value.
Solution: S+ = 50.00*1.25 = 62.50, c+ = 12.50, S– = 50.00*0.80 = 40.00, c– = 0.
h = (12.50 – 0)/(62.50 – 40.00) = 0.56.
c = hS + PV(–hS– + c–) = 0.56*50 + PV(-0.56*40 + 0) = 0.56*50 + (-0.56*40 + 0)/1.07
= 28.00 – 20.93 = 7.07.
Notice that buying a call option for $7.07 is equivalent to buying 0.56 units of the
underlying stock for $28.00. This purchase is partially financed (20.93) such that the
effective payment is $7.07.
Binomial Model: No Arbitrage Approach (Put Options)

For put options, the no-arbitrage single-period valuation equation is:
p = hS + PV(–hS– + p–) or equivalently,
p = hS + PV(–hS+ + p+)
where the hedge ratio h is given by the expression:

p+ − p−
h= +
S − S−
Based on these expressions, a put option can be interpreted as lending that is partially
financed with a short position in shares.
Example: A non-dividend-paying stock is currently trading at $50. A put option has one
year to mature, the periodically compounded risk-free interest rate is 7%, and the exercise
price is $50. Assume a single-period binomial option valuation model, where u = 1.25 and d
= 0.80. Estimate the option value.
Solution: S+ = 50.00*1.25 = 62.50, p+ = 0, S– = 50.00*0.80 = 40.00, p– = 10.
h = (0 – 10)/(62.50 – 40.00) = - 0.44.
p = hS + PV(–hS+ + p+) = - 0.44*50 + PV(- -0.44*62.5 + 0) = -0.44*50 + (0.44*62.5 + 0)/1.07
= -22.00 + 25.70 = 3.70.
Notice that buying a put option for $3.70 is equivalent to short selling 0.44 units of the
underlying stock for $22.00 and lending $25.70.
Binomial Model: Expectations Approach

The expectations approach is given by the following equations:
c = PV[πc+ + (1 – π)c–] and
p = PV[πp+ + (1 – π)p–]
π = the risk-neutral probability of an up move = (1 + r – d) / (u – d)
The expected terminal option payoffs can be expressed as:
E(c1) = πc+ + (1 – π)c– and
E(p1) = πp+ + (1 – π)p–
The option values can be written as:
c = PV[E(c1)] and
p = PV[E(p1)]
With the expectations approach,
 The probability, π, is objectively determined and is called the risk-neutral (RN)
probability. No assumption is made regarding the arbitrageur’s risk preferences.
 The discount rate is not risk-adjusted. It is simply the risk-free interest rate.
Example: A non-dividend-paying stock is currently trading at €100. A call option has one
year to mature, the periodically compounded risk-free interest rate is 5.15%, and the
exercise price is €100. Assume a single-period binomial option valuation model, where u =
1.35 and d = 0.74. What is the call option value and put option using the expectation
approach?
Solution: S+ = uS = 1.35(100) = 135

S– = dS = 0.74(100) = 74
c+ = Max(0,uS – X) = Max(0,135 – 100) = 35
c– = Max(0,dS – X) = Max(0,74 – 100) = 0
p+ = Max(0,100 – uS) = Max(0,100 – 135) = 0
p– = Max(0,100 – dS) = Max(0,100 – 74) = 26
π = [1 + r – d]/(u – d) = (1 + 0.0515 – 0.74)/(1.35 – 0.74) = 0.51
c = PV[πc+ + (1 – π)c–] = [(0.51)35 + (1 – 0.51)0] / 1.0515 = €17.00
p = PV[πp+ + (1 – π)p–] = [(0.51)0 + (1 – 0.51)26] / 1.0515 = €12.11
Two-Period Binomial Model

The two-period binomial lattice can be perceived as three one-period binomial lattices.
The option values under the two-period binomial model using the expectations approach
are:
c = PV[π2c++ + 2π(1 – π)c+– + (1 – π)2c– –]
p = PV[π2p++ + 2π(1 – π)p+– + (1 – π)2p– –]
American-Style Options
 American call options on non-dividend paying stocks should never be exercised early. If
the underlying stock is expected to decline in value, we can simply sell the option.
 On the other hand, it might make sense to exercise deep in-the-money American put
options and invest the sales proceeds at the risk-free rate.
 When early exercise has value, the no-arbitrage approach is the only way to value
American-style options. This approach requires that one works backward through the
binomial tree and determines whether early exercise has value at each step.
Example: Suppose you are given the following information: S0 = 26, X = 25, u = 1.466, d =
0.656, n = 2 (time steps), r = 2.05% (per period), and no dividends.
1. What is the value of a European put?
2. What is the value of an American put?

3. What is the early exercise premium of the American put?

Solution: To calculate the put values we should create the binominal tree as shown below:
We need to start at the end of the tree (Time 2) and work backwards. Node 2++ represents
two up moves of the underlying so the value is 26 x 1.466 x 1.466 = 55.88. The value of the
put option (whether European or American) is 0. Similarly:
Node 2+- (2-+): Underlying = 26 x 1.466 x 0.656 = 25. European put = American put = 0.
Node 2--: Underlying = 26 x 0.656 x 0.656 = 11.19. European put = American put = 0.
Node 1+: Underlying = 26 x 1.466 = 38.12. European put = American put = 0.
Node 1-: Underlying = 26 x 0.656 = 17.06. European put = 7.44, American put = 7.94.
At Node 1-, values of the European put and the American put are different. This is because
the American put can be exercised early but the European put cannot. The value of the
American put at Node 1- is X – S = 25 – 17.06 = 7.94. Since this value is more than the value
of the European put, it is beneficial to exercise the American put early.
The value of the European option is based on the option values at Nodes 2+- (2-+) and 2--,
the risk-neutral probability and the risk-free discount rate.
The RN probability, π = (1 + r – d)/(u – d) = (1 + 0.0205 – 0.656)/(1.466 – 0.656) = 0.45.
p- = PV[πp+ - + (1 – π)p- -] = (0 + 0.55 x 13.81) / 1.0205 = 7.44.
We now have the information to calculate the European put and American put value at
Node 0:
1. European put option = (0 + 0.55 x 7.44) / 1.0205 = 4.01.
2. American put option = (0 + 0.55 x 7.94) / 1.0205 = 4.28.
The early exercise premium = 4.28 – 4.01 = 0.27

Option on Interest Rates (Call)

 Interest rate options are options where the underlying is an interest rate.
 A call option on interest rates is in the money when the current spot rate is above the
exercise rate.
 A put option on interest rates is in the money when the current spot rate is below the
exercise rate.
 With interest rates, we assume that the risk neutral probability of an up move at each
node is 50%.
Example: Consider a two-year European-style call on the periodically compounded one-
year spot interest rate (the underlying). The spot rate is 3.0454%. Assume the notional
amount of the options is US$1,000,000 and the call exercise rate is 3.25% of par. Assume
the RN probability is 50%. The binomial interest rate tree is given below.
Rate 3.9706
Call 0.007206
Rate 3.9084
Discount
factor 0.962386
Call 0.003488
Rate 3.0454
Discount Rate 3.2542
factor 0.970446 Call 0.000042
Call 0.001702
Rate 2.6034
Discount
factor 0.974627
Call 0.000020
Rate 2.2593
Call 0
Note that the discount factor at each node depends on the interest rate at that node. Hence,
when the up move occurs at T = 1, the discount factor is based on an interest rate of
3.9084%.
At time step 2:
c++ = Max[0,3.9706% - 3.25%] = 0.007206. c+- = Max[0,3.2542% - 3.25%] = 0.000042
c-- = Max[0, 2.2593% - 3.25%]= 0
At time step 1:
c+ = PV1,2[πc++ + (1 – π)c+–]
= 0.962386[0.5(0.007206) + (1 – 0.5)0.000042] = 0.003488
c = PV1,2[πc+– + (1 – π)c– –]
–

= 0.974627[0.5(0.000042) + (1 – 0.5)0.0] = 0.00002

At time step 0:
c = PVrf,0,1[πc+ + (1 – π)c–]
= 0.970446[0.5(0.003488) + (1 – 0.5)0.00002] = 0.00170216
Since the notional amount is US$1,000,000, the call value is US$1,000,000(0.00170216)] =
US$1,702.16
Note: We can use a similar process to calculate the value of put options.
Black-Scholes-Merton Option Valuation Model

The BSM model is a continuous time version of the discrete time binomial model. It is used
to value path-independent options.
The assumptions of the BSM model are:
 The underlying follows geometric Brownian motion, implying a lognormal distribution
of the return, meaning that the continuously compounded logarithmic return is
normally distributed.
 Geometric Brownian motion implies continuous prices, meaning no jumps in the price
of underlying instrument only smooth movements from value to value.
 The underlying instrument is liquid, easily traded.
 Continuous trading is available, meaning able to trade at any moment.
 Short selling of the underlying instrument with full use of the proceeds is permitted.
 There are no market frictions, such as transaction costs, regulatory constraints, or
taxes.
 No-arbitrage opportunities available.
 All options are European-style, early exercise not allowed.
 The continuously compounded risk-free interest rate is known and constant; borrowing
and lending is at the risk-free rate.
 The volatility of the underlying return is known and constant.
 In case the underlying instrument pays a yield, it is expressed as a continuous known
and constant yield at an annualized rate.
The BSM Model

The inputs to the BSM model are:
 Price of underlying stock, S
 Continuously compounded risk-free rate, r
 Time to maturity in years, T
 Strike price, X
 Volatility of the underlying in annual percentage terms, σ
The BSM model for non-dividend paying stock is:

c = SN(d1) – e–rTXN(d2)
p = e–rTXN(–d2) – SN(–d1)
where:
ln(S/X) + (r + σ2 /2)T
d1 =
σ√T
d2 = d1 − σ√T
Using the above inputs, the BSM model can be used to predict:
 Call option price, c
 Put option price, p
The following tips will help you remember the formulas.
c = SN(d1) – e–rTXN(d2)
p = e–rTXN(–d2) – SN(–d1)
 A call option is of the form S- X, whereas a put option is of the form X – S.
 The present value of strike price X is obtained by multiplying it by e–rT.
 d1 is associated with S, whereas d2 is associated with X.
 For call options we use positive values of d1 and d2. Whereas, for put options we use
negative values of d1 and d2.
BSM Interpretations
The BSM model can be interpreted in multiple ways as indicated below.
Interpretation 1:
The BSM model can be interpreted as the present value of the expected option payoff at
expiration.
For calls c = PVr[E(cT)] and for puts p = PVr[E(pT)]
where:
E(cT) = SerTN(d1) – XN(d2) and E(pT) = XN(–d2) – SerTN(–d1).
Here e–rT is the present value term.
Interpretation 2:
The BSM model call value can be interpreted as a stock component, SN(d1), minus a bond
component, e–rTXN(d2). This is visible in the call option formula: c = SN(d1) – e–rTXN(d2).
The BSM model put value can be interpreted as a bond component, e–rTXN(–d2), minus a
stock component, SN(–d1). This is visible in the put option formula: p = e–rTXN(–d2) – SN(–
d1).
An option can be thought of as a dynamically managed portfolio of the underlying stock and
zero-coupon bonds. The price of the zero-coupon bond price, B = e–rTX.

Replicating Strategy Cost = nSS + nBB

 where the underlying shares nS = N(d1) > 0 for calls and nS = –N(–d1) < 0 for puts.
 The number of bonds is nB = –N(d2) < 0 for calls and nB = N(–d2) > 0 for puts.
For calls, the stock is bought because nS = N(d1) > 0 and the bond is sold because nB = –
N(d2) < 0. Selling a bond is the same as borrowing money. Therefore, a call option can be
viewed as a leveraged position in the stock.
For put options, the bond is bought because nB = N(–d2) > 0 and shares of the underlying
stock are sold because nS = –N(–d1) < 0. Buying a bond is the same as lending money. A put
can be viewed as buying a bond where this purchase is partially financed by short selling
the underlying stock.
The following table compares the BSM model with the Binomial model covered earlier.
Call option Put option
Underlying Financing Underlying Financing
Binomial model hs – –
PV(–hS + c ) hS PV(–hS– + p–)
BSM model N(d1)S –N(d2)e–rTX –N(–d1)S N(–d2)e–rTX
 N(d1) can be interpreted as the hedge ratio for call options.
 N(d2) can be interpreted as the probability that call option will expire in the money.
 1 – N(d2) = N(-d2) can be interpreted as the probability that put option will expire in the
money.
Example: Suppose we are given the following information on call and put options on a
stock: S = 100, X = 100, r = 5%, T = 1.0, and σ = 30%. Thus, based on the BSM model, it can
be demonstrated that PV(X) = 95.123, d1 = 0.317, d2 = 0.017, N(d1) = 0.624, N(d2) = 0.507,
N(–d1) = 0.376, N(–d2) = 0.493, c = 14.23, and p = 9.35.
1. Using the no-arbitrage approach, how can a call option be replicated?
2. Using the no-arbitrage approach, how can a put option be replicated?
Solution: The no-arbitrage approach to replicating the call option involves purchasing nS =
N(d1) = 0.624 shares of stock partially financed with nB = –N(d2) = –0.507 shares of zero-
coupon bonds priced at B = Xe –rT = 95.123 per bond. By definition, the cost of this
replicating strategy is nSS + nBB = 0.624(100) + (–0.507)95.123 = 14.17, which is the option
value.
The no-arbitrage approach to replicating the put option involves trading nS = –N(–d1) = –
0.376 shares of stock. The negative number means we need to short sell 0.376 shares. We
also need to buy nB = N(–d2) = 0.493 shares of the zero-coupon. Again, the cost of the
replicating strategy is nSS + nBB = –0.376(100) + (0.493)95.123 = 9.30 which is the value of
the put option.

BSM Model with Carry Benefits

The carry benefit-adjusted BSM model is:
c = Se–γTN(d1) – e–rTXN(d2)
and
p = e–rTXN(–d2) – Se–γTN(–d1)
where:
𝑙𝑛(𝑆/𝑋) + (𝑟 − 𝛾 + 𝜎2 /2)𝑇
d1 =
𝜎√𝑇
d2 = d1 – σ√T
The formulas for call and put options are similar to the formulas without carry benefit. The
stock price is adjusted downwards by multiplying it with e–γT. The bond component
remains unchanged.
Both d1 and d2 go down because of the adjustment made for carry benefits.
Foreign Exchange Options

Each currency option is for a certain quantity of foreign currency.
 The underlying is the foreign exchange spot rate.
 The underlying and the exercise price must be quoted in the same currency units.
 The carry benefit is the continuously compounded risk-free rate.
 Volatility is the volatility of the log return of the spot exchange rate.
The value of a currency call option is given by:
c = Se− rf TN(d1 ) - e–rTXN(d2)
Here S is the currency exchange rate in the price currency/base currency format, rf is risk-
free rate in the base currency and r is the risk-free rate in the price currency.
The BSM call model for currencies can be interpreted as the:
foreign exchange component, Se− rf TN(d1 ), minus the bond component, e–rTXN(d2).
The value of a currency put option is given by:
p = e–rTXN(–d2) - Se−rf TN(-d1)
The BSM put model for currencies can be interpreted as the:
bond component, e–rTXN(–d2), minus the foreign exchange component, Se−rf TN(-d1).
The formulas are similar to the formulas for stocks with dividends. Here, instead of
dividends, the benefit is the interest rate on the foreign currency. The spot rate S is

adjusted by multiplying it with e−rf T
European Options on Futures

The underlying instrument is a futures contract, and the assumption is that the futures
price follows geometric Brownian motion. Black’s model for European-style futures
options is as follows:
c = e–rT[F0(T)N(d1) – XN(d2)]
p = e–rT[XN(–d2) – F0(T)N(–d1)]
where:
ln[F0 (T)/X]+(σ2 /2)T
d1 =
σ√T
d2 = d1 − σ√T
Interpretation 1:
 c = present value of difference between futures price and exercise price adjusted by
N(d) functions.
 p = present value of difference between exercise price and futures price adjusted by
N(d) functions.
Interpretation 2:
The option value obtained through the Black model has two components, a futures
component and a bond component.
 c = futures component, F0(T)e–rTN(d1), minus the bond component, e–rTXN(d2)
 p = bond component, e–rTXN(–d2), minus the futures component, F0(T)e–rTN(–d1)
Interest Rate Options; Standard Market Model
With interest rate options the underlying is a forward interest rate. The standard market
model gives the present value of the expected option payoff at expiration.
c = (AP)e−rT [FRA(0, tj−1, tm) N(d1) − RXN(d2)]
p = (AP) e−rT [RXN(−d2) − FRA(0, tj−1, tm))N(−d1)]
where: AP = accrual period and Rx = exercise rate
Interpretation 1:
The standard market model can also be given as the present value of the expected option
payoff at expiration.
 c = PV[E(c)] where E(c) = (AP)[FRA(0, tj–1 ,tm)N(d1) – RXN(d2)]
 p = PV[E(p)] where E(p) = (AP)[RXN(–d2) – FRA(0,tj–1,tm)N(–d1)]
Interpretation 2:
The option value has two components, a futures component and a bond component.
 Call value = FRA component minus the bond component

 Put value = bond component minus the FRA component

The standard market model has the same general form as Black model, but there are
important differences. The differences with the standard market model are:
1. The discount factor, does not apply to the option expiration, tj–1 as in Black model
but applies to the maturity of the underlying FRA or tj (= tj –1 + tm).
2. The underlying is an interest rate (FRA), not a futures price.
3. The exercise price is an interest rate Rx, not a price.
4. The time to option expiration, tj–1, is used in the calculation of d1 and d2.
5. Finally, both the forward rate and the exercise rate should be expressed in decimal
form and not as percent. Or if expressed as a percent, then the notional amount
adjustment could be divided by 100.
Swaptions
There are two types of swaptions.
1. A payer swaption is an option on a swap to pay fixed, receive floating.
2. A receiver swaption is an option on a swap to receive fixed, pay floating.
The swaption model gives the present value of the expected option payoff at expiration.
PAYSWN = PV[E(PAYSWN,T)]
RECSWN = PV[E(RECSWN,T)]
Payer swaption = swap component – bond component

PAYSWN = (AP)PVA[RFIXN(d1) – RXN(d2)]
Receiver swaption = bond component – swap component

RECSWN = (AP)PVA[RXN(–d2) – RFIXN(–d1)]
The swaption model looks like the Black model but there are important differences:
1. The discount factor is absent. The payoff is not a single payment but a series of
payments. PVA (present value of an annuity) embeds the discount factor.
2. The underlying is the fixed rate on a forward interest rate swap.
3. The exercise price is an interest rate.
4. Both the forward swap rate and the exercise rate should be expressed in decimal
form and not as percent.
Option Greeks
Delta is defined as the change in a given portfolio for a given small change in the value of
the underlying instrument, holding everything else constant.
Gamma is defined as the change in a given portfolio delta for a given small change in the
value of the underlying instrument, holding everything else constant.

Theta is defined as the change in the value of an option given a small change in calendar
time, holding everything else constant.
Vega is defined as the change in a given portfolio for a given small change in volatility,
holding everything else constant.
Rho is defined as the change in a given portfolio for a given small change in the risk-free
interest rate, holding everything else constant.
Delta Hedging and Gamma Risk

Delta hedging refers to managing the portfolio delta by entering additional positions into
the portfolio. If DeltaH is the delta of the hedging instrument, the optimal number of units of
the hedging instruments, NH, is given by the formula below:
− Original Portfolio delta
NH = DeltaH
Example: Suppose we know S = 100, X = 100, r = 5%, T = 1.0, σ = 30%, and δ = 5%. We have
a short position in put options on 10,000 shares of stock. Based on this information, we
note Deltac = 0.532, and Deltap = –0.419. Assume each stock option contract is for one share
of stock.
1. Assuming the hedging instrument is a stock, how many stocks should be bought/sold
for a delta hedge?
2. Assuming the hedging instrument is a call option, how many call options should be
bought/sold for a delta hedge?
Solution to 1:
− Original Portfolio delta
B is correct. NH = DeltaH
The put delta is given as –0.419, thus the short put delta is 0.419. In this case, Portfolio
delta = 10,000(0.419) = 4,190 and DeltaH = 1.0. Thus, the number of hedging units is –4,190
[= –(4,190/1)] or short sell 4,190 shares of stock.
Solution to 2:
A is correct. Again, the Portfolio delta = 4,190 but now DeltaH = 0.532. Thus, the number of
hedging units is –7,875.9 [= –(4,190/0.532)] or sell 7,876 call options.
A few key points related to gamma are as follows:
 Gamma is always non-negative.
 Gamma has its largest value when an option is at the money.
 Gamma changes as the stock price changes and time to expiration change.
 Gamma approximates the estimation error in delta for options because the option
price with respect to the stock is non-linear.
 Gamma measures the non-linearity risk or the risk of the delta neutral portfolio.
 A gamma neutral portfolio means gamma is zero.

 When hedging an option portfolio, the option gamma has to be managed and then
the portfolio delta is neutralized.
Implied Volatility and Options Trading

 Implied volatility can be interpreted as the market’s view of how to value options.
 Given agreement on the pricing model, there is a one-to-one relationship between
implied volatility and option price.
 Implied volatility can be used to compare the values of two options, with different
exercise prices and expiration dates.
 If implied volatility is higher than expected volatility, the option is overpriced.
The BSM model assumes that volatility is known and constant and that all investors agree
on the value of volatility. In reality, there are can be different implied volatilities for calls
and puts with the same terms.
 Implied volatility can vary across exercise prices and the relationship is known as
the volatility smile or the skew depending on the particular shape.
 The implied volatility with respect to time to expiration is called the term structure
of volatility.
 A three-dimensional plot of the implied volatility with respect to both expiration
and exercise prices may be constructed which is known as the volatility surface.

Alternative Investments 2023 Level II High Yield Notes
LM01 Overview of Types of Real Estate Investment

Real estate investment: Basic forms
There are four basic forms of real estate investments: public equity, private equity, public
debt and private debt. Exhibit 2 presents some examples for each quadrant.
Public Private
Equity • Shares of REOCs • Direct investments in real estate,
• Shares of REITs, other listed including sole ownership and joint
trusts, exchange-traded funds ventures
(ETFs), and index funds • Indirect real estate ownership
through limited partnerships,
other forms of commingled funds,
or private REITS and REOCs
Debt • Mortgage REITs • Mortgages
• MBS (residential and • Private debt
commercial) • Bank debt
• Unsecured REIT debt
Each form has its own risks, expected returns, regulations, legal and market structures.
Investors can explore these characteristics and choose a quadrant that suits them.
 Private investments involve large investments and are illiquid. They also require
property management expertise.
 Public investments allow the ownership claim on a property to be divided. This
provides liquidity and diversification to the investors. Also, the properties are
professionally managed and no real estate management expertise is required on the
part of the investor.
 Equity investors take on more risk and therefore expect a higher rate of return than
debt investors. Their returns have two components: rent and appreciation of
property value.
 Debt investors get their returns from mortgage repayments and do not participate
in the appreciation of value of the underlying real estate.
Portfolio roles and economic value determinants of real estate investments
The motivations for investing in real estate are:

 Current income: Income generated by letting, leasing or renting the property.
 Price appreciation: Real estate prices rise over time.
 Inflation hedge: Both rents and property prices rise in an inflationary environment.
 Diversification: Low correlation with other asset classes such as stocks and bonds.
 Tax benefits: Some investors in certain countries may have potential tax benefits.

Exhibit 3 from the curriculum presents major economic factors that affect demand for the
major property types.
Commercial property types
The types of commercial properties are:

 Office – Demand depends on employment growth.
 Industrial and warehouse – Demand depends on overall strength of economy and
economic activity.
 Retail – Demand depends on trends in consumer spending, which in turn depends
on the health of the economy, job growth, population growth and savings rate.

 Multi-family – Demand depends on population growth and other demographics.
Due diligence for real estate investments
Due diligence of a property should include:

 Market review
 Lease and rent review
 Review costs of re-leasing space
 Look for supporting documentation
 Environmental inspection
 Physical/engineering inspection
 Review ownership history
 Review service and maintenance agreements
 Conduct property survey
 Verify property is compliant with zoning
 Verify property taxes, insurance and so on have been paid
Real estate investment indexes
The two main types of real estate investment indexes are:

Appraisal based indexes: These are based on appraisals. They suffer from appraisal lag
which underestimates volatility.
Transaction based indexes: These are based on actual transactions. The two main sub-
types are:
 Repeat sales index: Based on repeat sales of same property.
 Hedonic index: Does not require repeat sales of the same property, uses variables in
regression that control for differences.
Transaction based indexes are noisy and there may be random upward or downward
movements in the index.

LM02 Investments in Real Estate through Private Vehicles

Approaches to valuing real estate
Three major approaches are used to estimate real estate value:

 Income approach: Consider present value of expected future income from the
property, including resale at the end of a typical holding period. The concept is that
value depends on the expected rate of return that investors would require to invest
in the property.
 Cost approach: Consider what it would cost to buy land and construct a new
property that is similar to the subject property. The concept is that you should not
pay more for a property than the cost of buying vacant land and developing a
comparable property.
 Sales comparison approach: Consider transaction prices of comparable
properties. The concept is that you would not pay more than others are paying for
similar properties.
The value of a property can be estimated using all three approaches. The difference in the
estimates of value from each approach then needs to be reconciled to come up with a final
estimate of value of the subject property. This process is called ‘triangulation’.
The income approach to valuation
The income approaches used to estimate the value of a property are:

 Direct capitalization method: Estimates the value of an income-producing property
based on the level and quality of its net operating income.
 Discounted cash flow method: Discounts future projected cash flows to arrive at a
present value of the property.
In the direct capitalization method, value is based on capitalizing the first year NOI of the
property using a cap rate.
NOI1
Value =
Cap rate
Cap rate can be estimated from comparables:
NOI1
Cap rate =
comparable sales price
In discounted cash flow method, value is based on the present value of the property’s
future cash flows using an appropriate discount rate.
If NOI growth rate is constant:

NOI1
Value =
(r − g)
If NOI growth rate is not constant: 1) project NOI for a specific holding period 2) estimate
terminal value and 3) discount NOI and terminal value.
Financial ratios used to analyze real estate investments
Ratios and returns calculated and interpreted by debt and/or equity investors include:
 Loan-to-value ratio: Loan amount/ Appraised value of the property
 Debt service coverage ratio: NOI / Debt Service
 Equity dividend rate (cash-on-cash return): Measure of cash flow as a percentage of
equity investment
 Leveraged internal rate of return: IRR based on equity investment.
 Unleveraged internal rates of return: IRR based on total investment.

LM03 Investments in Real Estate Through Publicly Traded Securities

Types of REITs
Most REITs are structured as corporations or trusts. They are tax-efficient vehicles for
distributing earnings from rental income to shareholders.
A company must meet a number of criteria in order to qualify as a REIT. In most countries,
REITs must distribute 90%–100% of their otherwise taxable earnings, invest at least 75%
of their assets in real estate, and derive at least 75% of their income from real estate rental
income or mortgage interest.
The main types of REITs are:
 Equity REITs: Have ownership position in income-producing real estate.
 Mortgage REITs: Invest the bulk of their assets in loans secured by real estate.
Net asset value approach for REIT valuation
Net asset value per share (NAVPS) is the difference between a REIT’s assets and its
liabilities (all taken at current market values rather than accounting book values), divided
by the number of shares outstanding.
NAVPS = (Value of assets – Value of liabilities) / number of shares outstanding
The actual share price can be different from NAVPS. Shares can trade either at a discount or
a premium to NAVPS. NAVPS is the largest component of the intrinsic value of the stock.
NAVPS calculation
The current market value of real estate assets is calculated by capitalizing NOI. The market
values of other assets and liabilities are assumed to be equal to their book values.
Exhibit 2 from the curriculum provides an example of NAVPS calculation. It starts with the
net operating income (NOI) of a property and shows the various adjustments which need
to be made.
Office Equity REIT Inc. Net Asset Value Per Share Estimate
(In Thousands, Except Per Share Data)
Last 12-months real estate NOI $270,432
Less: Non-cash rents 7,667
Plus: Adjustment for full impact of acquisitions
4,534
(1)
Pro forma cash NOI for last 12 months $267,299
Plus: Next 12 months growth in NOI (2) $4,009
Estimated next 12 months cash NOI $271,308
Assumed cap rate (3) 7.00%

Office Equity REIT Inc. Net Asset Value Per Share Estimate
(In Thousands, Except Per Share Data)
Estimated value of operating real estate $3,875,829
Plus: Cash and equivalents $65,554
Plus: Land held for future development 34,566
Plus: Accounts receivable 45,667
Plus: Prepaid/Other assets (4) 23,456
Estimated gross asset value $4,045,072
Less: Total debt $1,010,988
Less: Other liabilities 119,886
Net asset value $2,914,198
Shares outstanding 55,689
(1) 50 percent of the expected return on acquisitions was made in the middle of 2010.
(2) Growth is estimated at 1.5 percent.
(3) Cap rate is based on recent comparable transactions in the property market.
(4) This figure does not include intangible assets.
NAVPS is calculated to be $2,914,198 divided by 55,689 shares, which equals $52.33 per
share.
Relative value approach for REIT valuation
Funds from operation (FFO) is accounting net earnings excluding:

1. Depreciation charges on real estate
2. Deferred tax charges (deferred portion of tax expenses)
3. Gains/losses from sale of property and debt restructuring
Adjusted funds from operation (AFFO) is a refinement to FFO and is designed to be a
more accurate measure of current economic income.
AFFO = FFO – straight line adjustment – recurring maintenance type capital expenditures
and leasing commissions
In relative value approach, we use the P/FFO, P/AFFO, EV/EBITDA multiples to value
REITs. The main drivers behind these multiples are:
1. Expectations for growth in FFO
2. Risk associated with underlying real estate
3. Risk associated with capital structure and access to capital
P/FFO and P/AFFO Multiples: advantages and drawbacks
Advantages:
1. These multiples are widely accepted.

2. Portfolio managers can put REIT valuations into context with other investment
alternatives. This makes it easier to compare REITs with other investments.
3. FFO estimates are readily available.
4. Multiples can be used in conjunction with growth rates and leverage levels for
relative value analysis.
Drawbacks:
1. Does not capture the intrinsic value of all real estate assets. For example, empty
buildings do not contribute to FFO but have value.
2. P/FFO does not adjust for recurring capital expenditures. Although P/AFFO
considers this, there are wide variations in estimates and assumptions.
3. One-time gains/losses create issues with this model.
Private vs. Public real estate investments
Exhibit 9 summarizes some of the key differences, advantages, and disadvantages of public
and private real estate investing.



LM04 Private Equity Valuation

Sources of value creation
The sources of value creation in private equity are:

1. The ability to re-engineer the firm to generate superior returns.
2. The ability to access credit markets on favorable terms.
3. Better alignment of economic interests of private equity owners and managers of
portfolio companies.
Alignment of interests
Private equity firms are able to achieve better alignment of interests by:
 Allowing managers to focus on the long term perspective, as compared to short term
quarterly earnings targets in public companies.
 Effective structuring of investment terms:
o Tag-along, drag-along rights: any future acquirer has to extend acquisition offer
to all shareholders, including management of the company.
o Corporate board seats: ensure private equity control in case of a major corporate
event.
o Non-compete clause: prevents founders from restarting same activity during a
predefined period of time.
o Preferred dividend and liquidation preference: private equity firms are paid
first, before other shareholders.
o Reserved matters: some domains of strategic importance are subject to approval
by private equity firm.
o Earn-outs: a mechanism linking the acquisition price paid by the private equity
firm to the company’s future financial performance over a predetermined time
horizon.
Valuation characteristics of buyout vs. venture capital investments
Buyout investments: Venture capital investments:

Steady and predictable cash flows Low cash flow predictability, cash flow
projections may not be realistic
Excellent market position (can be a niche Lack of market history, new market and
player) possibly unproven future market (early
stage venture)
Significant asset base (may serve as basis Weak asset base
for collateral lending)

Strong and experienced management team Newly formed management team with
strong individual track record as
entrepreneurs
Extensive use of leverage consisting of a Primarily equity funded. Use of leverage is
large proportion of senior debt and rare and very limited
significant layer of junior and/or
mezzanine debt
Risk is measurable (mature businesses, Assessment of risk is difficult because of
long operating history) new technologies, new markets, lack of
operating history
Predictable exit (secondary buyout, sale to Exit difficult to anticipate (IPO, trade sale,
a strategic buyer, IPO) secondary venture sale)
Established products Technological breakthrough but route to
market yet to be proven
Potential for restructuring and cost Significant cash burn rate required to
reduction ensure company development and
commercial viability
Low working capital requirement Expanding capital requirement if in the
growth phase
Buyout firm typically conducts full blown Venture capital firm tends to conduct
due diligence approach before investing in primarily a technology and commercial due
the target firm (financial, strategic, diligence before investing; financial due
commercial, legal, tax, environmental) diligence is limited as portfolio companies
have no or very little operating history
Buyout firm monitors cash flow Venture capital firm monitors achievement
management, strategic, and business of milestones defined in business plan and
planning growth management
Returns of investment portfolios are Returns of investment portfolios are
generally characterized by lower variance generally characterized by very high
across returns from underlying returns from a limited number of highly
investments; bankruptcies are rare events successful investments and a significant
number of write-offs from low performing
investments or failures
Large buyout firms are generally Venture capital firms tend to be much less
significant players in capital markets active in capital markets
Most transactions are auctions, involving Many transactions are “proprietary,” being
multiple potential acquirers the result of relationships between venture
capitalists and entrepreneurs

Strong performing buyout firms tend to Venture capital firms tend to be less
have a better ability to raise larger funds scalable relative to buyout firms; the
after they have successfully raised their increase in size of subsequent funds tend
first funds to be less significant
Variable revenue to the general partner Carried interest (participation in profits) is
(GP) at buyout firms generally comprise generally the main source of variable
the following three sources: carried revenue to the general partner at venture
interest, transaction fees, and monitoring capital firms; transaction and monitoring
fees fees are rare in practice
LBO model and VC method for valuation
LBO model:
Value is created in a typical leveraged buyout transaction from a combination of factors
such as:
 Earnings growth due to operational improvements and improved corporate
governance
 Multiple expansion depending on pre-identified potential exits
 Debt reduction, repayment of part of debt with operational cash flows before the
exit.
Exhibit 4 from the curriculum shows a typical leveraged buyout value creation chart.
Leverage is very important in buyout transactions. As debt is gradually paid off, a larger
proportion of operating cash flows is available to equity investors. However, these high
levels of debt significantly increase the risk to equity investors. This increased risk should
be taken into consideration, when comparing the returns with other classes such as stock
market.
VC method:
Two important concepts in the VC method are:
 Pre-money valuation: Agreed value of a company prior to a round of financing
 Post-money valuation: Value of a company after the financing or investing round

The relationship between the two is:

Pre-money valuation + New equity investment = Post-money valuation.
The proportionate ownership of the venture capital investor is equal to Investment/Post-
money valuation.
VC firms generally target an expected hurdle rate. Using the expected hurdle rate and the
projected value of the company at exit, we can calculate the post money valuation as:
Value of equity at exit
Post money valuation =
ROI
where:
ROI = (1 + IRR)t
The pre-money valuation can then be calculated as:
Pre-money valuation = Post-money valuation – New equity injection
Example:
Consider an entrepreneur who is looking to raise $500,000. Given the size of its market and
the industry, the entrepreneur’s company expects to reach sales of $80 million over the
investment horizon. A typical revenue multiple for a revenue-generating business in its
industry is 2×. If the VC firm’s ROI is 20× and the entrepreneur’s company has no debt, then
what is the pre-money valuation? And what is the VC firm’s fractional ownership?
Solution:
Expected exit valuation = performance metrics x industry multiple = $80 million x 2 = $160
million
Value of equity at exit $160
Post money valuation = ROI
= 20
= $8 million
Pre-money valuation = Post-money valuation – New equity injection = $8 million - $0.5

million = $7.5 million
VC fractional ownership = $0.5 million / $8 million = 6.25%
Exit routes their impact on value
Private equity investors have the following exit routes for their investments:
 Initial public offering (IPO): The key points are:
o Highest exit value relative to other methods
o High liquidity, access to capital, and attracts good management
o Less flexible, more costly, and complex
o Used when company has strong growth prospects, operating history, size
o Timing of IPO is an important consideration
 Secondary market: Sale to other financial investors or strategic investors. The

advantages are
o Highest value in absence of an IPO
o Bring portfolio companies to next level by restructuring, merger, new market
etc. and sell them to a strategic investor or to other PE firm.
 Management buyout: Firm is sold to management. Best alignment of interest,
however, if significant leverage is used it can reduce the company’s flexibility.
 Liquidation: Sale of firm’s assets. This option is used if the company is no longer
viable.
Private equity fund structures, terms, valuation and due diligence
Fund structure:
Most PE firms are structured as limited partnerships, where the fund manager is the
general partner (GP) and the fund’s investors are limited partners (LP). The GP has
management control over the fund and is jointly liable for all debts. The LPs have limited
liability; they do not risk more than the amount of their investment in the fund.
Two core functions of the GP are: (1) to raise funds and (2) To manage investments
Exhibit 6 from the curriculum shows the funding stages for a private equity firm.
Terms:
The most significant economic terms are:
 Management fees: represent a percentage of committed capital paid annually to the
GP.
 Transaction fees: fees paid to GPs in their advisory capacity when they provide
investment banking services for a transaction.

 Carried interest: represents the general partner’s share of profits generated by a

private equity fund.
 Ratchet: mechanism that determines the allocation of equity between shareholders
and the management team of the portfolio company.
 Hurdle rate: the IRR that a private equity fund must achieve before the GP receives
any carried interest.
 Target fund size: expressed as an absolute amount in the fund prospectus.
 Vintage year: the year the private equity fund was launched. Reference to vintage
year allows performance comparison of funds of the same stage and industry focus.
 Term of the fund: typically 10 years, extendable for additional shorter periods (by
agreement with the investors)
The most significant corporate governance terms are:
 Key man clause: If a key executive leaves, the GP is prohibited from making new
investments until a new key executive is appointed.
 Disclosure and confidentiality: Private equity firms have no obligations to disclose
publicly their financial performance. Some terms limit the information available to
investors.
 Clawback provision: If a fund makes profitable exits in early years, but the
subsequent exits are less profitable, then the GP has to pay back profits to ensure
that the profit split is in line with the fund prospectus. The two types of clawback
provisions are:
o Due on termination
o Annual reconciliation (true-up)
 Distribution waterfall: Mechanism to ensure LPs are paid first before the GP
receives carried interest. The two main types are:
o Deal-by-deal waterfall – Allowing distribution after each deal.
o Total return waterfall – Distribution is calculated on the entire portfolio.
 Tag-along, drag along rights: Any future acquirer has to extend acquisition offer to
all shareholders, including the management of the company.
 No-fault divorce: A GP may be removed without cause, if a super majority of LPs
approve that removal.
 Removal for “cause”: Allows removal of GP or early termination of fund for causes
such as gross negligence, ‘key person’ event, felony conviction, bankruptcy, or
material breach of the fund prospectus.
 Investment restrictions: Minimum level of geographic or sector diversification, or
limits on borrowing.
 Co-investment: LPs have the first right of co-investing along with the GP if the GP
launches a new fund.

Valuation:
The value of a fund is based on NAV. The fund’s assets are valued by GPs in the following
ways:
1. At cost with significant adjustments for subsequent financing events or
deterioration.
2. At lower of cost or market value.
3. By a revaluation of a portfolio company whenever a new financing round involving
new investors takes place.
4. At cost with no interim adjustment until the exit.
5. With a discount for restricted securities.
6. More rarely, marked to market by reference to a peer group of public comparables
and applying illiquidity discounts.
Due diligence:
Due diligence is important because:
 PE funds show strong persistence of returns over time. This means that top
performing funds tend to continue to outperform and poor performing funds also
tend to continue to perform poorly or disappear.
 Difference between performances of funds is extremely large. For example, the
difference between top quartile and third quartile fund IRRs can be about 20
percentage points.
 Liquidity in private equity is low and LPs are locked for the long term. On the other
hand, when private equity funds exit an investment, they return the cash to the
investors immediately. Therefore, the “duration” of an investment in private equity
is typically shorter than the maximum life of the fund.
Risks and costs of investing in private equity
The risks are:

 Illiquidity of investments
 Unquoted investments
 Competition for attractive investment opportunities
 Reliance on management of investee companies
 Loss of capital
 Government regulations
 Taxation risk
 Valuation of investments
 Lack of investment capital
 Lack of diversification
 Market risk

The costs are:

 Transaction fees
 Investment vehicle fund setup costs
 Administrative costs
 Audit costs
 Management and performance fees
 Dilution
 Placement fees
Evaluating fund performance
Analysis of private equity fund’s financial performance includes the following:

 Gross IRR: Relates to cash flows between the fund and its portfolio companies. It is
considered a good measure of the investment management team’s track record in
creating value.
 Net IRR: Relates to cash flows between the fund and LP’s. It measures the returns to
investors.
 PIC (paid in capital): The ratio of paid in capital to date divided by committed
capital.
 DPI (distributed to paid in): Cumulative distributions paid out to LPs as a
proportion of the cumulative invested capital. DPI is presented net of management
fees and carried interest.
 RVPI (residual value to paid in): Value of LPs’ shareholding held with the private
equity fund as a proportion of the cumulative invested capital. RVPI is presented net
of management fees and carried interest.
 TVPI (total value to paid in): The portfolio companies’ distributed and undistributed
value as a proportion of the cumulative invested capital. TVPI is the sum of DPI and
RVPI. TVPI is presented net of management fees and carried interest.

LM05 Introduction to Commodity and Commodity Derivatives

Characteristics of commodity sectors
The following table summarizes the characteristics of six important commodity sectors:
Sector Description Storage/ Supply Demand
Transport
Energy Crude oil, refined Natural storage, Political events, Economic growth
products, natural pipes, ships new
gas technologies
Grains Corn, wheat, rice, Easy Weather, Humans, animal
soy; seasons disease, pests feed, fuel
Industrial Copper, Storage easy; Not impacted by Industrial growth
metals aluminum, nickel, transport can weather
zinc, lead, tin, iron be expensive
Livestock Poultry, sheep, Linked to grain Grain costs, Emerging markets
cattle, hogs costs weather, disease
Precious Gold, silver, Easy Not impacted by Inflation,
metals platinum weather technology,
jewelry
Softs Cotton, coffee, Difficult - Weather Wealth, emerging
(cash sugar, cocoa freshness is markets
crops) important
Commodity sectors life cycle
The following table summarizes the life cycle for six important commodity sectors (Copper
is used as an example for industrial metals; and coffee is used as an example for cash
crops):
Commodity Steps Seasons Considerations Contracts
Crude oil Input-output Extracted year Refineries and Futures

production life cycle: round; seasonal pipelines are contracts and
extraction, demand based very expensive indexes which
transportation, on weather. to build but follow local
storage, trading, these costs are grades and
refining, much lower than origins.
transportation and the cost of
trading exploration.

Natural gas Straight-through

consumption:
extraction,
transportation,
storage, trading
Copper Input-output Extracted year Economies of Futures
production life cycle: round. scale; difficult to contracts come
extract, grind, reduce supply due every month
concentrate, roast, when demand is of the year.
smelt, convert, refine, low.
store and transport
Livestock Time to maturity Growth is year- Historically US Ranchers and
increases with size round; weight livestock exports slaughterhouses
Cattle: birth, feeder gain is are low because trade futures to
cattle, live cattle, influenced by of high risk of hedge exposure.
slaughter. weather and spoilage; now
pastures. this risk is lower
because of
advances in
cryogenics.
Grains Planting, growth, Well defined Stored in silos Farmers and
pod/ear/head seasons and and warehouses. consumers trade
formation, harvest. growth cycles. futures to hedge
Demand is year- exposure.
round.
Coffee Harvested somewhere year-round. Two major Two futures
Cycle: Plant, three to four years to bear varieties: contracts:
fruit (cherry), harvesting is done by robusta and robusta in
hand in multiple sweeps, two to three arabica. Brazil London and
weeks to dry, hulled, sorted and bagged produces both. arabica in New
for final market. Local buyer roasts and York.
ships to retail location.
Valuation of commodities
 Stocks and bonds are financial assets that represent claims against future cash flows.
They are generally priced based on the present value of cash flows.
 On the other hand, commodities generally do not generate a stream of cash flows,
therefore their valuation cannot be done on the basis of cash flows. Instead, the value of
a commodity is the discounted value of a future price. The future price depends on

factors such as supply, demand and expected volatility.

 Investment in commodities is primarily done via derivative instruments such as futures
contracts. The price of commodity futures contracts is based on the current spot price
as well as the transportation and storage cost of holding a commodity.
 Future contracts can either be cash settled or require physical delivery. Market
participants may not always have the ability to make or take physical delivery, and this
factor can affect future prices.
 Hedgers can also impact futures prices as they regularly trade in order to hedge their
commodity exposures.
Futures market participants
Commodity market participants include:

 Hedgers – want to hedge their exposure to a commodity.
 Traders and investors
o Informed investors – take advantage of mispricing due to a lack of information
in the market place.
o Liquidity providers - trade with hedgers and effectively provide insurance
services.
o Arbitrageurs – seek to profit from mispricing between the futures price and
spot price.
 Commodity exchanges - determine contracts, rules and procedures for commodity
trading and act to ensure that market participants fulfill obligations.
 Commodity market analysts - perform research and analysis on commodities and
recommend investment positions.
 Commodity regulators – responsible for regulation of commodity markets.
Spot and futures pricing
Spot price is the current price of a commodity for immediate delivery at a specific location.
Spot prices vary across different regions based on quality and local supply demand factors.
Futures price is the price for future delivery of a commodity of a specific quality at a
specific location.
The difference between the spot price and futures price is called ‘basis’.
 When spot price is higher than futures price, the market is in backwardation and
basis is positive.
 When the spot price is lower than the futures price, the market is in contango and
basis is negative.
The price difference between two contracts of different maturities is called the 'calendar

spread'.
 Calendar spread = Price of contract maturing earlier - Price of contract maturing
later.
 Calendar spread is positive when the futures market is in backwardation and
negative when the futures market is in contango.
Commodity futures contracts can be either cash settled or require physical delivery.
Theories explaining futures returns
There are three theories of commodity futures returns:

1. Insurance theory: Commodity producers sell the commodity in futures market in
order to make their revenues predictable. This implies that the futures price must
be less than the spot price as a payment to the speculator for providing insurance to
the producer.
2. Hedging pressure hypothesis: Producers and consumers sell and buy
(respectively) in the futures market in order to remove price uncertainty. The
number of sellers as against the number of buyers would determine whether the
futures market would be in contango or backwardation.
3. Theory of storage: Commodity futures prices are affected by the costs of storage
and convenience yield. Convenience yield is inversely related to general available of
the commodity. Futures price = Spot price + Direct Storage costs - Convenience yield
Components of futures returns
The total return of a fully collateralized commodity futures contract is made up of:
1. Price return is produced by a change in spot prices.
Price return = (Current spot price - Previous spot price)/Previous spot price
2. Roll return is produced by closing expiring contracts and reestablishing the
position in far-dated contracts.
Roll return is sector dependent, it is positive when futures markets are in
backwardation and negative when futures markets are in contango.
Roll return can be significant for a single period but is a small percentage of total
return over multiple periods.
Gross roll return = (Near-term contract closing price – Farther-term contract closing
price) / Near-term futures contract closing price.
Net roll return = Gross roll return × Percentage of the position in the futures
contract being rolled.
3. Collateral return is the yield on securities that the investor deposits as collateral to
establish the futures position.

Collateral return = Risk free rate return

Total return = price return + roll return + collateral return.
Commodity swaps
A commodity swap is a legal contract involving the exchange of payments over multiple
dates as determined by specified reference prices or indexes relating to commodities.
Swaps offer the participant greater customization relative to a futures contract. They also
offer the advantage of providing the participant with a series of futures contracts without
him having to actually manage multiple contracts.
Types of swaps:
 Excess return: party pays a premium and receives excess return over a strike price.
 Total return: party receives total return on a commodity or index.
 Basis: payments are based on two related commodity reference prices.
 Variance: variance buyer benefits if actual variance is higher than stated variance
and the direction of price move matters.
 Volatility: similar to variance swap except based on volatility and the direction of
price moves does not matter.
Commodity indexes - Key characteristics
Commodity indexes portray the aggregate movement of a basket of commodities. Key

characteristics of a commodity index include:
 Breadth of coverage - Number of commodities and sectors included in the index.
 Weights - Relative weights assigned to each commodity and the methodology for
determining these weights.
o Production weighted: a commodity sector’s weight is based on the level of
production or economic value of that sector. There could be a cap or floor on
the weight of a particular sector or commodity. In an upward trending
market production value weighted indexes will outperform.
o Equal weighted: all sectors have the same weight.
o Liquidity weighted: liquidity can also be considered in allocating the weight
of a sector.
 Rolling methodology: determining how contracts that are about to expire will be
rolled over into future months.
o A passive roll methodology typically involves owning ‘front contracts’.
o An active roll methodology would try to maximize roll yield.
 Rebalancing methodology: the methodology and frequency of rebalancing the
weights of individual commodities, sectors and contracts in the index.
 Governance: The process by which above mentioned characteristics are

implemented.
In general, commodity indexes have a low correlation with traditional asset classes such as
stocks and bonds. However, the correlation across different commodity indexes (or futures
indexes) tends to be high.

Portfolio Management 2023 Level II High Yield Notes
LM01 Exchange-Trade Funds: Mechanics and Applications

The creation/redemption process
ETFs rely on a creation/redemption process that is carried out in an OTC primary market
between the ETF issuer and authorized participants (a special group of institutional
investors).
 The AP creates new ETF shares by transacting in-kind with the ETF issuer: A pre-
specified basket of securities is exchanged for a certain number of shares in the ETF.
At the start of every business day, the ETF manager publishes a list of required in-
kind securities. This list is called the creation basket.
 The process also works in reverse: The AP can present the ETF shares to the ETF
issuer for redemption and it will receive the basket of underlying securities in
return.
The creation/redemption mechanism rewards the AP for keeping the price of the ETF in a
tight range around the NAV of basket securities. Due to market conditions, the prices of the
ETF and basket securities change continuously. The AP monitors both to identify profitable
arbitrage opportunities.
An advantage of the ETF creation/redemption process is that the AP absorbs all transaction
costs. These costs are then passed on to investors in the secondary market through the bid-
ask spread. Therefore, frequent ETF traders bear the cost of their trading activity. Buy-and-
hold ETF shareholders are not affected by the negative impact of transaction costs caused
by other investors entering and exiting the fund. This makes the ETF structure inherently
more fair as compared to mutual funds (where transaction costs affect all investors).
As the creation and redemption process happens in kind, it allows the ETF’s portfolio
manager to manage the cost basis of their holdings, which leads to greater tax efficiency.
ETFs - secondary markets

ETFs trade on both the primary market (between the ETF issuer and APs) and on
secondary markets. End investors trade ETFs with each other on the secondary market.
ETFs go through the same settlement and clearing process as other listed stocks.
Sources of tracking error for ETFs

Although the expense ratio of an ETF is a useful measure, it does not fully reflect the cost of
holding an ETF. To understand the true costs, we need to evaluate how well an ETF tracks
its underlying index.
While evaluating how well an ETF tracks its underlying index, a rolling return assessment
using periodic performance deviations (tracking differences) is more useful than the
standard deviation of daily returns (tracking error).

Factors responsible for ETF tracking differences are:

 Fees and expenses
 Representative sampling/optimization
 Use of depositary receipts and other ETFs
 Index changes
 Fund accounting practices
 Regulatory and tax requirements
 Asset manager operations
Factors affecting ETF bid–ask spreads

ETF BAS vary by trade size. They are usually published for smaller trade sizes. For large
trade sizes, the BAS is small, and it can also be negotiated directly, therefore it is not
published. BAS are tightest for ETFs that are very liquid and have continuous two-way
order flow.
Factors affecting ETF bid-ask spreads include:
 Creation/redemption costs and other direct trading costs, such as brokerage and
exchange fees.
 Bid-ask spread of underlying securities.
 Risk of hedging or carrying positions by liquidity provider.
 Market maker’s target profit spread. This depends on the level of competition in the
ETF market.
 Discount related to likelihood of receiving offsetting ETF order.
The BAS of ETFs holding international stocks are tightest when the underlying security
markets are open for trading.
Fixed income ETFs tend to have wider BAS than equity ETFs. This is because the
underlying bonds trade in dealer markets; hence the BAS for the underlying bonds are
wider and hedging is more difficult.
Sources of ETF premiums and discounts to NAV

ETF premiums and discounts refer to the differences between the exchange price of the
ETF and the fund’s calculated NAV.
The major sources of ETF premiums and discounts are:
 Timing difference: NAV is usually a poor value indicator for ETFs holding foreign
securities because of differences in exchange closing times.
 Stale pricing: Premiums and discounts can occur because NAVs are based on the last
traded prices, which may be observed at a time lag to the ETF price, or because the
ETF is more liquid and better reflects the current supply and demand than the
underlying securities.

Costs of owning an ETF

The costs of owning an ETF may be positive or negative and implicit or explicit. Also, some
costs depend on the holding period, while others are independent of the holding period.
The main components of ETF cost are:
 Fund management fee
 Tracking error
 Portfolio turnover
 Trading costs, such as commissions, bid–ask spreads, and premiums/discounts
 Taxable gains/losses
 Security lending
Trading costs are a more significant consideration for shorter-term tactical ETF investors.
Whereas, management fees are a more significant consideration for longer-term buy-and-
hold investors.
Types of ETF risk

Risks of investing in ETFs include:
 Counterparty risk: Exchange-traded notes (ETNs) are subject to default by the ETN
issuer. ETFs which use OTC derivatives such as swaps face settlement risk. ETFs
may lend their securities for additional income, which exposes them to the risk of
counterparty default.
 Fund closures: ETF closures can create unexpected tax liabilities.
 Expectation-related risks: Some ETFs provide access to complex asset classes and
strategies. An investor may not fully understand the underlying exposure.
ETFs in portfolio management

Primary ETF strategies include:
 Portfolio efficiency: The use of ETFs to better manage a portfolio for efficiency or
operational purposes. The applications include:
o transacting cash flows for benchmark exposure
o rebalancing to target asset class or risk factor weights
o filling exposure gaps in portfolio holdings of other strategies and funds
o temporarily holding during transitions of strategies or managers
 Asset class exposure management: The use of ETFs to achieve or maintain core
exposure to key asset classes, market segments, or investment themes on a
strategic, tactical, or dynamic basis.
 Active and factor investing: The use of ETFs to target specific active or factor
exposures on the basis of an investment view or risk management need.

LM02 Using Multifactor Models

Arbitrage pricing theory (APT)
APT is a multifactor model that describes the expected return on an asset as a linear
function of the risk of the asset with respect to a set of factors. It is an alternative to the
capital asset pricing model (CAPM). But, unlike the CAPM, APT does not identify or specify
the number of risk factors.
The generic form of a multifactor model is given by:
R i = ai + bi1 F1 + bi2 F2 + ⋯ + biK FK + εi
Where ai is the intercept, Fi are the factors and bi1 is the sensitivity to factor 1.
The three major assumptions of APT are:
 A factor model describes asset returns.
 There are many assets, so investors can form well-diversified portfolios that
eliminate asset-specific risk.
 No arbitrage opportunities exist among well-diversified portfolios.
General case of the APT equation
The generic form of the APT equation is given below. It says that the expected return on
any well-diversified portfolio is linearly related to the factor sensitivities of that portfolio.
E(R p ) = R F + λ1 βp ,1 + ⋯ + λK βp ,K
where:
E(R p ) = expected return to portfolio p
R F = the risk − free rate
λj = the expected reward for bearing the risk of factor j or the factor risk premium
βp ,j = the sensitivity of the portfolio to factor j
K = the number of factors
If the portfolio has a sensitivity of 1 to factor 1 and 0 to all other factors, then it is called a
pure factor portfolio. The return in this case is given by:
E(R p ) = R F + λ1
Let us take an example. If the expected return of the portfolio is 12% and the risk-free rate
is 4%, what is the risk premium for Factor 1?
12 = 4 + 𝜆1 ; 𝜆1 = 8%
An arbitrage opportunity is an opportunity to earn an expected positive net profit
without risk and with no net investment of money. An arbitrage opportunity exists if a
portfolio offers either a high or low expected return relative to another portfolio with the

same factor sensitivity.
Factors and types of multifactor models
A factor is a common or underlying element with which several variables are correlated.
Systematic factors are factors that affect the average returns of a large number of
different assets. They represent priced risk.
There are three types of multifactor models: macroeconomic factor models, fundamental
factor models, and statistical factor models.
Macroeconomic factor models: the factors are surprises in macroeconomic variables that
significantly explain equity returns.
Fundamental factor models: the factors are attributes of stocks or companies that are
important in explaining cross-sectional differences in stock prices.
Statistical factor models: statistical methods are applied to a set of historical returns to
determine factors that explain historical returns in one of two senses:
 In factor analysis models, the factors are the portfolios that best explain (reproduce)
historical return covariances.
 In principal-component models, the factors are portfolios that best explain
(reproduce) the historical return variances.
The advantage of statistical models is that they use minimal assumptions. But the
disadvantage is that they are difficult to interpret relative to the other two models.
The structure of macroeconomic factor models
Macroeconomic models are based on the principle that the returns to an asset are
correlated with surprises in macroeconomic factors such as GDP, inflation, etc. A factor
surprise is the actual value minus predicted value. Factor surprises are the independent
variables in this model. The generic form of a macroeconomic factor model is given below:
R i = ai + bi1 F1 + bi2 F2 + ⋯ + biK FK + εi
Consider a factor model in which the returns to a stock are correlated with two
macroeconomic factors: surprises in inflation rates and surprises in GDP growth. The
return to stock i is modeled as:
R i = ai + bi1 FINT + bi2 FGDP + εi
where:
R i = the return to stock i
ai = the expected return to stock i
bi1 = the sensitivity of the return to stock i to interest rate surprises
FINT = the surprise in interest rates

bi2 = the sensitivity of the return to stock i to GDP growth surprises

FGDP = the surprise in GDP growth
εi = an error term with a zero mean that represents the portion of the return to asset I not
explained by the factor model
A factor sensitivity is a measure of the response of return to each unit of increase in a
factor, holding all other factors constant.
The structure of fundamental factor models
Unlike with macroeconomic models, the factors in fundamental factor models represent
returns instead of surprises. This means that the factors do not have expected values of
zero and the intercept is not the expected return to asset i.
An example of the fundamental factor model is the Carhart four-factor model:
E(R p ) = R F + βp ,1 RMRF + βp ,2 SMB + βp ,3 HML + βp ,4 WML
where:
RMRF = the return on a value-weighted equity index in excess of the one-month T-bill rate.
This is the market risk factor.
SMB = small minus big, a size (market capitalization factor); SMB is the average return on
three large-cap portfolios. This is the size factor.
HML = high minus low, the average return on two high book-to-market portfolios minus
the average return on two low book-to-market portfolios. This is the value factor.
WML = winners minus losers, a momentum factor; WML is the return on a portfolio of the
past year’s winners minus the return on a portfolio of the past year’s losers. This is the
momentum factor.
Macroeconomic vs. Fundamental factor models
The table below summarizes the differences between macroeconomic factor models and
fundamental factor models:
Macroeconomic factor models Fundamental factor models
Factor stated as a surprise relative to what Factor stated as a return.
was expected.
Sensitivity of the stock’s return to a Factor sensitivities are attributes of the
surprise in a macroeconomic variable (ex: security; an asset’s sensitivity to a factor is
GDP growth surprises) expressed using a standardized beta.
Develop factor (surprise) series first and Generally, specify the factor sensitivities
then estimate the sensitivities through (attributes) first and then estimate the
regression. factor returns through regression.

Fixed income multifactor models
Of the three types of multifactor models (macroeconomic, fundamental, and statistical),

statistical models can be most easily applied to various asset classes, including fixed
income. However, macroeconomic and fundamental models require adjustments and
repurposing before they can be applied to fixed income.
Macroeconomic Multifactor Models
Macroeconomic factors such as surprises to economic growth, interest rates, and inflation
impact bond pricing. The return to bond i, Ri, can be modeled as:
Ri = ai + bi1FINFL + bi2FGDP + ϵi,
where:
Ri = the return to bond i
ai = the expected return to bond i
bi1 = the sensitivity of the return on bond i to inflation rate surprises
FINFL = the surprise in inflation rates
bi2 = the sensitivity of the return on bond i to GDP growth surprises
FGDP = the surprise in GDP growth (assumed to be uncorrelated with FINFL)
ϵi = an error term with a zero mean that represents the portion of the return to bond i
not explained by the factor model
Fundamental Multifactor Models
Fundamental factor approaches address the unique aspects of fixed income by using
categories such as:
 Duration: ranging from cash to long-dated bonds
 Credit: ranging from government securities to high yield
 Currency: ranging from home currency to foreign developed and emerging market
currencies
 Geography: specific developed and emerging markets
The return to bond i, Ri, can be modeled as:
Ri = ai + bi1FGvt_Sh + bi2FGvt_Int + bi3FGvt_Lg + bi4FInvest + bi5FHiYld + bi6FMBS + ϵi,
where:
Ri = the return to bond i
ai = the expected return to bond i
bik = the sensitivity of the return on bond i to factor k
Fk = factor k, where k represents “Gov’t (Short),” “Gov’t (Long),” and so on
ϵi = an error term with a zero mean that represents the portion of the return to bond i
not explained by the factor model

Factor models in return attribution
Active managers try to generate a return above a given benchmark by holding securities in
weights different than the benchmark. The active return on a portfolio is the return on a
portfolio minus the benchmark’s return.
Active return = R P − R B
Multifactor models help us understand the sources of a manager’s return relative to a
benchmark. Analysts favor fundamental multifactor models to decompose the sources of
returns as they are easier to explain compared to statistical models.
Using a factor model, we can decompose a portfolio manager’s active return as the sum of
two components:
 Return from factor tilts: product of the portfolio manager’s factor tilts (overweight
or underweight relative to the benchmark factor sensitivities) and the factor
returns. For example, if a benchmark’s sensitivity to market risk is 1.0 and an active
portfolio’s sensitivity to market risk 1.2, the market risk factor tilt here is 1.2 – 1.0 =
0.2.
 Return from asset selection: part of active return reflecting the manager’s skill in
individual asset selection. That is, overweighting securities that outperform the
benchmark or underweighting securities that underperform the benchmark.
The following equation shows the decomposition of active return into these two
components:
Active return =
∑Kj=1[(Portfolio sensitivity)j − (Benchmark sensitivity)j ] ∗ (Factor return)j +
Asset selection
Factor models in risk attribution
Active risk, also known as tracking error or tracking risk, is the standard deviation of
active returns. Tracking error is expressed as:
TE = s(R p − R B )
where s(R p − R B ) is the sample standard deviation of the time series of differences
between the portfolio return and the benchmark return.
The information ratio is used to measure active returns per unit of active risk. The
formula for IR is given by:
Active return ̅p − R
R ̅B
IR = =
Active risk s(R p − R B )
For example, consider a portfolio that has a sample mean return of 9% and the benchmark

has a sample mean return of 7.5%. If the portfolio’s tracking error is 6%, then what is its
IR?
IR = (9 − 7.5)/6 = 0.25
If we use variances rather than standard deviation to measure risk, then the risk measure is
called active risk squared. The formula for active risk squared is given as:
Active risk squared = s 2 (R p − R B )
Active risk squared = Active factor risk + Active specific risk
Active factor risk is the risk due to portfolio’s different-than-benchmark exposures relative
to factors specified in the risk model.
Active specific risk are risks resulting from the portfolio’s active weights on individual
assets. It is also known as asset selection risk. This is the active non-factor or residual
risk.
Factor models in portfolio construction
Multifactor models are used in portfolio construction. They allow portfolio managers to
make focused bets and control portfolio risk relative to the benchmark’s risk.
Using multifactor models to model a portfolio’s risk is useful in:
 Passive management: portfolio managers of a fund tracking an index with many
securities often select a set of sample securities from the index. They then use factor
models to replicate the index’s factor exposures with this sample.
 Active management: multifactor models are used to predict alpha.
 Rules-based active management: rules/automation are used to replicate active
management. The focus of portfolio construction in this approach is to tilt towards
factors such as size, quality, value or momentum. The rules-based approach
attempts to capture systematic exposure at a low-cost.
A factor portfolio is a portfolio with a sensitivity of 1 to a particular factor and a sensitivity
of 0 to all other factors.
How factor considerations can be useful in strategic portfolio decisions
Multifactor models help investors in the following ways:

 Investors can make strategic investment decisions, by taking on risk factors that the
investor has a comparative advantage in bearing and avoiding risk factors that the
investor has a comparative disadvantage in bearing.
o For example, a university endowment might have a comparative advantage in
bearing business cycle risk of traded equities or liquidity risk or private equity
investment because of their long investment horizon. Since these risks do not

affect them much, they may increase the asset allocation within an asset class to
capture the risk premium for these risks.
o Similarly, an investor who depends on income from salary will be more sensitive
to business cycle risk relative to an independently wealthy investor.
 Investors can achieve efficient and better-diversified portfolios.

LM03 Measuring and Managing Market Risk

Value at risk
Value at risk (VaR) is the minimum loss that would be expected to be incurred for a certain
percentage of the time over a certain period of time given assumed market conditions.
Consider a $400 million portfolio. The VaR statement for this portfolio might be as follows:
“The 5% VaR is $2.2 million over a one-day period.” Notice that the VaR measure has three
important elements:
 Minimum loss - In our example, the minimum expected loss is $2.2 million. The
minimum loss can be expressed in either currency units or percentage terms.
 Time horizon - In our example, the time horizon is one day. This is the time period
over which the minimum loss can be expected.
 Frequency – In our example, the frequency is 5%. This can be interpreted as ‘a loss
of $2.2 million or more is expected 5% of the time’ or ‘95% of the time, the loss will
be less than $2.2 million.’
There are three methods to estimate VaR:
1. Parametric (variance - covariance) method
2. Historical simulation method
3. Monte Carlo simulation method
Parametric method
The parametric method is also known as the variance-covariance method. The detailed
steps for calculating VaR using this method are:
1. Break down the portfolio into a set of risk factors. Simplistically, each security in the
portfolio could represent a risk factor.
2. Model the return distribution of the risk factors and the portfolio (typically a normal
distribution).
3. Calculate the expected return of the portfolio over the given time period. If VaR is to
be estimated over a one day period, then the single day expected return would be
calculated.
4. Calculate the volatility of the portfolio over the given time period. If VaR is to be
estimated over a one day period, then the single day volatility would be calculated.
5. Estimate the VaR by finding the point in the distribution that lies to the left of the
mean as per the defined threshold.
Advantages:
 It is simple and straight-forward.
 The assumption of normal distribution allows for an easy estimation of VaR.

Disadvantages:
 The normal distribution cannot be applied in all cases.
 The VaR under this method is very sensitive to parameter estimates; especially
covariance.
 It cannot be applied to a portfolio that contains options, as options have a non-
normal return distribution.
Historical simulation method
The historical simulation method uses the current portfolio and re-prices it using the actual
historical returns of the key risk exposures observed in the lookback period. The steps
involved are:
1. Break down the portfolio into a set of risk factors.
2. Gather the historical returns of each factor for the chosen lookback period.
3. Reprice the current portfolio based on the actual returns on each day of the
lookback period.
4. Sort the portfolio returns starting from the largest loss and going up to the largest
gain.
5. Choose the point prior to which the number of outcomes correspond with the
defined VaR threshold; for instance, a 5% VaR over one day would mean we choose
the outcome beyond which 5% of the outcomes result in larger losses.
Advantages:
 Estimated outcomes are based on actual historical data; hence, results cannot be
dismissed as impossible.
 This method can be adopted for portfolios with options, as the method is based on
actual historical returns.
Disadvantages:
 There is no certainty that historical outcomes will re-occur, therefore, the estimated
results may not be relevant for the future.
Monte Carlo simulation method
Under the Monte Carlo simulation method of VaR estimation, the user develops his own set
of assumptions regarding the statistical characteristics of the return distribution and based
on these characteristics, generates random outcomes of portfolio returns. The steps
involved in this method are as follows:
1. Break down the portfolio into a set of risk exposures.
2. Develop assumptions regarding the return distributions of the risk exposures.
3. Generate a set of return outcomes based on the assumptions for each risk exposure.
4. Reprice the portfolio based on the return outcomes generated.

5. Sort the results from the largest loss to the largest gain.
6. Choose the point prior to which the number of outcomes correspond with the
defined VaR threshold. A 5% Monte Carlo VaR would simply be the fifth percentile
of the simulated values instead.
Advantages:
 Avoids complexity inherent in the parametric method when the portfolio has a large
number of assets, as a large number of assets makes it difficult to specify the
parameters of the distribution.
 Does not need to be constrained by the assumption of normal distribution.
 Can handle portfolios with options.
Disadvantages:
 There is no set industry standard for the number of random values to be generated.
Generating more random values can make results more reliable but can be more
time-consuming.
 Generating random values of multiple exposures requires specifying their
correlation, which must be accounted for.
 Quality of output depends on the quality of input. If the input distributions and
parameters are not correctly specified, the output will be meaningless.
Note: The parametric method and the Monte Carlo method give the same VaR if
parameters and distributions used are the same and the number of simulations is very
large.
Advantages and limitations of VaR
The advantages of VaR are:

 It is a simple concept and is relatively easy to understand.
 It is easily communicated, as a lot of information is captured in a single number.
 It can be useful in comparing risks across asset classes, portfolios, and trading units
and, as such, it facilitates capital allocation decisions.
 It can be used for performance evaluation.
 It can be verified through backtesting.
 It is widely accepted by regulators.
The primary limitations of VaR are:
 It is a subjective measure and highly sensitive to numerous discretionary choices
made in the course of computation.
 It can underestimate the frequency of extreme events.
 It fails to account for the lack of liquidity.
 It is sensitive to correlation risk.

 It is vulnerable to trending or volatility regimes.

 It is often misunderstood as a worst-case scenario.
 It can oversimplify the picture of risk.
 It focuses heavily on the left tail.
Extensions of VaR
Conditional VaR is the average loss conditional on exceeding the VaR cutoff.
Incremental VaR measures the change in portfolio VaR as a result of adding or deleting a
position from the portfolio or if a position size is changed relative to the remaining
positions.
MVaR measures the change in portfolio VaR given a small change in the portfolio position.
Relative VaR measures the degree to which the performance of a given investment
portfolio might deviate from its benchmark
Sensitivity risk measures
Sensitivity measures determine how portfolio performance changes with respect to

changes in a single risk factor. The different types of sensitivity risk measures are:
 Equity risk is measured by beta.
o Beta: Sensitivity of the asset’s return to the market risk premium.
 Interest rate risk of fixed-income securities is measured by duration and convexity.
o Duration: Sensitivity of the bond’s price to changes in its yield.
o Convexity: A second-order effect which measures changes in duration.
 Option risk is measured by delta, gamma and vega.
o Delta: Sensitivity of option price to the price of the underlying
o Gamma: A second-order effect which measures changes in delta
o Vega: Sensitivity of option price to its volatility
Scenario risk measures
Scenario risk measures estimate how portfolio performance would be affected when either
a hypothetically created scenario or a historical scenario is applied to the current portfolio.
This involves changing multiple risk factors in the scenario and determining the outcome
on portfolio performance.
 Historical scenarios measure the hypothetical portfolio return that would result
from a repeat of a particular period of financial market history.
 Hypothetical scenarios are extreme movements and co-movements that have not
occurred before. They utilize a technique known as reverse stress testing where

significant risk exposures of the portfolio are first determined and then these
exposures are tested in changing scenarios.
Single-factor stress tests are a sub-category of scenario analysis. These tests involve
making a substantial change to a factor and evaluating the impact on the portfolio. These
tests are also called ‘pass/fail stress tests’ because they help determine whether an entity
will survive a major shock.
Limitations of scenario analysis
 Historical scenarios are unlikely to reoccur in the exact same way.

 Hypothetical scenarios do not use historical asset movements as a reference,
therefore they are likely to model asset correlations or magnitude of movements
incorrectly or unrealistically.
 Hypothetical scenarios can be difficult to create; modeling asset movements and
correlations with reasonable accuracy is demanding and time consuming.
 Extreme scenarios can be difficult to justify as realistic. This results in a bias among
scenario creators to underestimate movements in order to appear credible.
Sensitivity and scenario risk measures and VaR
Sensitivity risk measures, scenario risk measures and VaR offer a different picture of a
portfolio’s risk exposures.
VaR vs. Sensitivity risk measures:
 VaR measures losses and the probability of large losses. In contrast, sensitivity risk
measures show changes in value of an asset given changes in other parameters but
does not assign a probability to such changes.
 VaR gives a broad picture in that it measures an estimated minimum loss but does
not identify where the loss may come from; hence, it is not useful for identifying
sources of risk. In contrast, sensitivity risk measures allow a detailed look into the
relationships driving the risk exposures of a portfolio and are useful for identifying
sources of risk.
VaR vs. Scenario risk measures:
 VaR and scenario risk measures are similar in that they both measure a potential
loss.
 VaR and hypothetical scenario risk measures are different because VaR uses
historical returns as inputs. Hence, if historical asset correlations don’t hold in
future, VaR would be less useful. In contrast, scenario risk measures allow the risk
assessment to be hypothetical and thus relax the biases imposed by VaR.

Applications of risk measures
Three factors typically determine the types of risk measures used by different market
participants.
 The degree to which the market participant is leveraged.
 The mix of risk factors to which their business is exposed.
 The accounting or regulatory requirements governing their reporting.
Banks use risk tools to assess the extent of any liquidity and asset/liability mismatch, the
probability of losses in their investment portfolios, their overall leverage ratio, interest rate
sensitivities, and the risk to economic capital.
Asset managers use risk tools to evaluate volatility, probability of loss, or the probability
of underperforming a benchmark.
Pension funds use risk measures to evaluate asset/liability mismatch and surplus at risk.
Property and casualty insurers use sensitivity and exposure measures to ensure
exposures remain within defined asset allocation ranges, economic capital and VaR
measures to estimate the impairment in the event of a catastrophic loss, and scenario
analysis to test the market risks and insurance risks simultaneously.
Life insurers use risk measures to assess the exposures of the investment portfolio and the
annuity liability, the extent of any asset/liability mismatch, and the potential stress losses
based on the differences between the assets in which they have invested and the liabilities
resulting from the insurance contracts they have written.
Using constraints in market risk management
Constraints are widely used in risk management in the form of risk budgets, position limits,
scenario limits, stop-loss limits, and capital allocation.
 Risk budgeting is the allocation of the total risk appetite across sub-portfolios.
 A scenario limit is a limit on the estimated loss for a given scenario, which, if
exceeded, would require corrective action in the portfolio.
 A stop-loss limit requires a reduction in the size of a portfolio, or its complete
liquidation, when a loss of a particular size occurs in a specified period.
 Position limits are limits on the market value of any given investment.
Note: Some firms might use marginal VaR for each trading desk. This means allocating each
desk a VaR budget such that the total VaR is the sum of each individual desk’s marginal
VaR. This approach permits each trading desk to “reinvest” the diversification benefits
obtained at the aggregate level.

Risk measures and capital allocation
Capital allocation under risk management is done to place limits on each of a company’s
activities in order to ensure that the areas in which it expects the greatest reward are given
the resources needed to accomplish their goals.
The economic capital of a business is the amount of capital it needs to hold in order to
survive losses from risks inherent in its business. It is the minimum amount of capital that
a company must maintain. The actual balance sheet capital should exceed the economic
capital. Once the overall economic capital of the business is defined, it is then allocated
among its business activities.
Capital allocation is popular among:
 entities which face significant tail risk.
 entities which use significant leverage.

LM04 Backtesting and Simulation

Introduction and objectives of backtesting
Four commonly used methods to evaluate investment strategies are:

 Backtesting: Tests a strategy in an actual historical environment, usually over a
long period of time. Tells us how a strategy would have performed if it had been
implemented in the past.
 Historical scenario analysis: Tests a strategy in a specific historical environment,
usually over a short period of time, such as during recessions or periods of high
inflation.
 Simulation: Rather than actual historical environments, simulation tests a strategy
in a hypothetical environment specified by the user.
 Sensitivity analysis: Determines how changes in input variables affect a target
variable and risk profile.
Objectives of backtesting: Backtesting approximates the real-world investment process
by evaluating whether a strategy would have produced desirable results using historical
data. It can be used to decide if an investment strategy should be accepted or rejected.
The backtesting process
Backtesting consists of three steps: strategy design, historical investment simulation, and
analysis of backtesting output.
Step 1: Strategy design

In this step we identify the investment goals and hypothesis. Typical goals for active
strategies are:

 To achieve excess returns over a benchmark

 To achieve superior risk-adjusted absolute return
An investment hypothesis is a method designed for achieving the goal. Key parameters
defined under an investment hypothesis include:
 Investment universe: This term refers to all of the securities in which we can
potentially invest. The constituents of well-known broad market indexes (for
example, the Russell 3000 Index) are frequently used as the investment universe.
 Return definition: If the investment universe spans multiple countries, we must
decide which currency to calculate the return in. Two popular choices are:
o Translate all investment returns into the home country currency
o Denominate returns in local currencies
If the investment strategy’s goal is to achieve excess returns over a benchmark, a
suitable benchmark must also be specified.
 Rebalancing frequency and transaction cost: Rebalancing frequency refers to how

often a portfolio is updated to reflect current data. It is common to use a monthly
rebalancing frequency.
 Start and end date: Typically, investment managers prefer to backtest investment
strategies over as long a period of time as possible, because a larger sample size
provides greater statistical confidence in the results.
Step 2: Historical investment simulation
Analysts typically use ‘rolling windows’ to simulate historical investment. In this approach
a portfolio is constructed at the start of a period using data from a historical in-sample
period, followed by testing on a subsequent, out-of-sample period. The process is repeated
as time moves forward.
Step 3: Analysis of backtesting output
The final step in backtesting is generating results for presentation and interpretation. We
are concerned not only with the portfolio’s average return but also with its risk profile.
Therefore, metrics such as the Sharpe ratio, the Sortino ratio, volatility and maximum
drawdown are used to evaluate performance. Apart from these measures, visuals are also
frequently used – for example, the distributions of returns can be plotted against a well-
known distribution, such as the normal distribution.
Common problems in backtesting
Common problems in a backtest of an investment strategy include:

 Survivorship bias: Refers to drawing conclusions from data that only includes
entities that have survived to that point. It is a common mistake that investors make

when conducting back tests using current index constituents. This bias can be
avoided by using point in time data.
 Look ahead bias: Created by using information that was unknown or unavailable
during the historical periods over which the backtest is conducted. Look ahead bias
can occur due to – reporting lags, revisions of macroeconomic data, and index
additions. This bias can be avoided by using point in time data.
 Data snooping: Refers to making an inference after looking at statistical results
rather than testing a prior inference. The bias occurs when an analyst selects data or
performs analyses repeatedly until a significant result is found. This bias can be
avoided by using a high hurdle rate or by using cross validation.
Historical scenario analysis
Historical scenario analysis is a type of backtesting used to assess the performance and risk
of an investment strategy in different structural regimes and at structural breaks.
Two common examples of regime changes are:
 Expansions and recessions
 High-and low-volatility regimes
Simulation analysis
Backtesting implicitly assumes that the past is likely to repeat itself. However, this
assumption may not hold true. Simulation analysis is used to supplement backtesting and
to obtain a more complete picture.
The two basic types of simulation are:
 Historical simulation: Instead of assuming that a strategy was implemented at
some past date and collecting results as the strategy runs over time, we construct
results by randomly selecting returns from many different historical periods
without regard to time-ordering.
Thus, as compared to backtesting we relax a key constraint by randomly changing
the sequence of historical periods from which factor returns are drawn.
Historical simulation is widely used in investment management, mainly by banks for
market risk analysis.
 Monte Carlo simulation: The issue with historical simulation is that there is only
one set of realized data to draw from (in other words the past happened in only one
way). Monte Carlo simulation helps overcome this issue. Here each key variable is
assigned a statistical distribution, and observations are drawn at random from that
distribution.
Thus, the main advantage of Monte Carlos simulation is that the analyst does not

have to rely on historical distributions, he can specify any distribution and

incorporate non-normality, fat tails, tail dependence, and so on, to model key
variables. But the downside is that the method is complex and computationally
intensive.
A simulation is implemented using the following eight steps:
1. Define the target variable. Typically, this variable is the return on an investment
strategy and its distribution.
2. Specify key decision variables. Typically, these variables are the returns of each
underlying asset and their weights in the overall portfolio.
3. Specify the number of trials (N) to run. Typically, researchers choose between 1,000
and 10,000 simulation runs. The greater the number of simulations the more
reliable the results.
4. Define the distributional properties of the key decision variables.
 In historical simulation, we draw from historical data.
 In Monte Carlo simulation, we can select and specify an appropriate statistical
distribution for each key decision variable.
5. Use a random number generator to draw N random numbers for each key decision
variable.
6. For each set of simulated key decision variables, compute the value of the target
variable.
7. Repeat steps 5 and 6 ‘N’ times.
8. We now get a set of N values of the target variable. We can then compute typical
metrics like mean return, volatility, Sharpe ratio, and the various downside risk
metrics such as CVaR and maximum drawdown.
Sensitivity analysis
Sensitivity analysis is a technique for assessing how changes in input variables affect a
target variable and risk profiles. It can be implemented to help managers better understand
the potential risks and returns of their investment strategies.
Sensitivity analysis frequently uses the multivariate skewed Student’s t-distribution which
can account for skewness and excess kurtosis. But it requires estimation of more
parameters and is thus more likely to suffer from larger estimation errors.
The procedure is similar to simulation. Steps 4 and 5 are the only exception. In Step 4,
instead of fitting the data to a multivariate normal distribution, we calibrate our model to a
multivariate skewed t-distribution. In Step 5, we simulate 1,000 sets of factor returns from
this new distribution function.

LM05 Economics and Investment Markets

Framework for the economic analysis of financial markets
According to the present value model, the current market price of an asset is the sum of the
asset’s expected future cash flows discounted at an appropriate risk adjusted discount rate.
The equation for the present value model is as follows:
N
̃ t+s
Et [CF i
]
Pti =∑ s
s=1
(1 + lt,s + θt,s + ρit,s )
The denominator represents the discount rate, which can be decomposed into three
components:
 Real risk free interest rate, lt,s
 Expected inflation, θt,s
 Risk-premium for uncertainty in future cash flows, ρit,s
Two observations can be made from the present value model equation:
 Asset values are based on expectations.
 Any new information that is different from current expectations will affect asset
values. For example, if the expected earnings growth rate was 5%, but the real
growth rate turns out to be 4%, then this will bring down the asset’s value.
The real default-free interest rates
The price today of an investment in a zero-coupon bond at time t that is certain to pay off
one unit of real consumption in s periods is given by:
Pt,s = Et [1m
̃ t,s ] = Et [m
̃ t,s ]
In the above expression, mt,s is the investor’s marginal willingness to trade consumption at
time t for (real) wealth at time t + s. For example, if mt,s is 0.9, then this implies that the
investor is willing to give up $0.9 today to have $1 of real consumption at the end of period
1. mt,s is also called the ‘inter-temporal rate of substitution’.
The inter-temporal rate of substitution will be higher if the economy is expected to perform
poorly. Investor’s will be willing to reduce current consumption and save more (e.g. 0.95)
to have 1 unit of real consumption later.
On the other hand, if the economy is expected to be good, the inter-temporal rate of
substitution will be low. Investor’s will be willing to give up less, e.g. 0.85, to have 1 unit of
real consumption later.
If there is a 0.5 probability of a poor economy and a 0.5 probability of a good economy, the
expected value of the inter-temporal rate of substitution is given by:

Et [m
̃ t,1 ] = 0.5 × 0.85 + 0.5 × 0.95 = 0.9
If an investor’s inter-temporal rate of substitution (e.g. 0.92) is higher than the current
bond price (0.90), then she will keep buying the bond until her inter-temporal rate of
substitution becomes equal to the bond price.
The one-period risk free rate is calculated as:
1 − Pt,1
lt,s =
Pt,1
If uncertainty about the state of the economy increases the expected inter-temporal rate of
substitution will be lower, i.e., investors will demand higher return when uncertainty
increases.
Risk premiums on risky assets
Consider a default-free bond with a maturity of two periods that pays 1 unit of real income
after two periods. Assume the bond is held for only one period. The price at the end of
period 1 is uncertain because the bond is sold before its maturity date. This price
uncertainty gives rise to a risk premium, although the bond is default-risk free. Given this
information, the price of the bond at time 0 can be calculated as:
̃t+1,s−1 m
Pt,s = Et [P ̃ t,1 ]
The curriculum derives the relationship between expected values and covariance. From an
exam perspective, it will suffice to remember that the pricing equation above can be
equated to the following:
̃t+1,s−1]
Et [P
̃t+1,s−1 m
Pt,s = Et [P ̃ t,1 ] = ̃t+1,s−1 , m
+ covt [P ̃ t,1 ]
1 + lt,1
where:
̃t+1,s−1 , m
cov𝑡 [P ̃ t,1 ] is the covariance between an investor’s inter-temporal rate of
substitution, m˜t,1, and the random future price of the investment at t + 1, ̃ Pt+1,s−1
Covariance for risky assets is expected to be negative because:
 When the expected future price of the investment is high, the marginal utility of
future consumption relative to current consumption is low.
 If expected labor income is high, the marginal utility of future consumption relative
to current consumption is low.
Negative covariance implies that asset prices are lower, which results in a positive risk
premium. The larger the negative covariance term, the higher the risk premium and
required return of an asset will be.

Default-free interest rates and economic growth
There are two important points that relate change in economic growth to real default-free
interest rates:
 Higher trend economic growth results in higher real default-free interest rates.
 Higher volatility of GDP growth rate results in higher real default-free interest rates.
For a real default-free bond, the price can be expressed as:
N i
CFt+s
Pti =∑ s
s=1
(1 + lt,s )
As you can see from the above equation, the only factor that will affect the price of the bond
is lt,s , which is determined by the real economic growth and the volatility in economic
growth over time. The aggregation of the consumption and saving decision of individuals
decides these economic factors.
Short-term rates and the business cycle
The price of a default-free nominal coupon-paying bond can be expressed as:

N i
CFt+s
Pti =∑ s
s=1
(1 + lt,s + θt,s + πt,s )
where:
θt,s = additional return required by investors to compensate for inflation
πt,s = risk premium for uncertainty in future inflation
With short-term nominal interest rates, inflation uncertainty can be ignored and short-
term T-bills can be priced as:
i
CFt+s
Pti = s
(1 + lt,s + θt,s )
Short-term nominal interest rates are positively related to short-term real interest rates
and short-term inflation expectations. The interest rates are higher in economies with
higher inflation and higher, more volatile growth.
Central banks set short-term interest rates in response to the economy’s position in the
business cycle.
 They cut rates when economic activity and/or inflation is slow.
 They increase rates when economic activity and/or inflation is high.
The Taylor rule is a rule for setting policy rates and determining whether the rate is at an

appropriate level. The rule is given below:

prt = lt + it + 0.5(it − i∗t ) + 0.5(Yt − Yt∗ ) = lt + 1.5it − 0.5i∗t + 0.5(Yt − Yt∗ )
where:
prt = the policy rate at time t
lt = the level of real short-term interest rates that balance long-term savings and borrowing
in the economy
𝑖𝑡 = the rate of inflation
𝑖𝑡∗ = the target rate of inflation
Yt and Yt∗ = the logarithmic levels of actual and potential real GDP respectively.
A neutral policy rate is the policy rate that neither spurs nor impedes real economic
activity.
The default-free yield curve and the business cycle
Break-even inflation = Yield on a zero-coupon default-free nominal bond – yield on a zero-

coupon default-free real bond with same maturity.
The break-even rate considers the inflation expectations of investors over the investment
horizon and a risk premium for the uncertainty about future inflation.
The evolution of nominal default-free bonds (i.e. the yield curve) is impacted by the
following:
 Inter-temporal rate of substitution.
 Changing expectations about inflation.
 Changing perceptions about the uncertainty of future inflation environments
Government bond risk premiums:
 are positive.
 are probably related to the consumption hedging benefit of government bonds.
 are positively related to bond maturity.
 fall during bad economic times.
An approximate method for calculating risk premiums on default free bonds is:
Risk premium = Yield on nominal bond – yield on real bond – inflation expectation
Yield curves
The following exhibit reproduced from the curriculum show the UK government bond yield
rates for July 2007, December 2010, and December 2011

From the yield curve, it is evident that the UK short-term interest rates were higher in 2007
than the long-term interest rates. There was a significant change (decline) in interest rates
between July 2007 and December 2010.
Yield curves have three characteristics:
 Level: views of future inflation determine the level.
 Slope: the magnitude of the risk premium determines the slope. The slope increases
during a recession.
 Curvature
Most of the yield curve movement can be explained by changes in level, followed by slope,
followed by curvature.
The shape of the yield curve is determined by:
 Expectations of real short-term interest rates: views of future economic activity
drive short-term rates.
 Inflation expectations: if investors expect inflation and interest rates to be higher in
the future, it results in an upward sloping yield curve.
 Risk premiums: for bonds with longer maturity, the uncertainty in future inflation is
higher which requires a higher risk premium.
 Other factors such as regulations requiring banks to hold more short-term T-bills, or
pension funds to hold long-term bonds.
The term spread and the business cycle
A yield curve is downward sloping when investors expect inflation and future interest rates
to decline. An inverted yield curve is a sign of recession.
When the economy is in a recession and interest rates are low, the yield curve will be
upward sloping. A booming economy has high inflation and future short-term interest

rates.
Credit premiums and the business cycle
The pricing formula for bonds with credit risk is given by:
N
̃i t+s ]
Et [CF
Pti =∑ s
s=1
(1 + lt,s + θt,s + πt,s + γit,s )
Credit premium is represented by γit,s . It is the additional return required by investors as

compensation for bearing credit risk.
Credit spread = yield on credit-sensitive bond – yield on government bond
Some important points related to credit spreads are summarized below:
 Credit spreads increase in times of economic weakness. The probability of default
increases and recovery rate decreases when the economy slows down. Therefore,
investors demand a return (yield) for possible losses.
 On the other hand, credit spreads narrow as the economy strengthens.
 When spreads narrow, credit-sensitive bonds perform well.
 Low-rated bonds tend to outperform when the economy is strengthening.
Conversely, when the economy weakens, bonds with a high credit-rating tend to
outperform bonds with lower ratings as the spread between the issuers widens.
Apart from the business cycle, there are other factors that affect the credit spread of a
corporate bond:
 Issuer’s industrial sector and rating. Credit spreads of low-rated bonds or those
issued by companies in the cyclical sector widen the most.
 Company specific factors such as profitable issuers with low debt interest payments
and not dependent on debt financing have a high credit rating. This is because their
ability to pay is relatively high.
Credit risk premiums for sovereign bonds are based on issuer’s ability to pay and the
likelihood that they might default.
Instructor’s note: The core point to remember is that credit risky bonds perform poorly in
bad economic times.
Equities and the equity risk premium
The pricing equation or equities can be expressed as follows:

∞
̃i t+s ]
Et [CF
Pti = ∑
i 𝑖 s
s=1
(1 + lt,s + θ t,s + πt,s + γ t,s + 𝜅𝑡,𝑠 )

𝑖
This equation is similar to the pricing equation for credit-risky bonds except 𝜅𝑡,𝑠 , which
represents the premium demanded by investors for investing in equities relative to bonds.
The equity risk premium can be assumed to be a sum of a risk premium for investing in
𝑖
equities, 𝜅𝑡,𝑠 , and a premium for holding a default-free government bond, γit,s .
𝑖
Equity risk premium = λit,s = 𝜅𝑡,𝑠 + γit,s
Therefore, the above equation can also be expressed as:
∞
̃i t+s ]
Et [CF
i
Pt = ∑ s
s=1
(1 + lt,s + θt,s + πt,s + 𝜆𝑖𝑡,𝑠 )
Equities do not have good consumption hedging properties, i.e., they do not pay off well
during bad economic times. Therefore, risk averse investors would demand a higher
premium on equity holdings relative to bond holdings.
Equity risk premium and credit risk premium are positively correlated and influenced by
the business cycle in similar ways. But equities are more sensitive to the business cycle
than credit-risky bonds.
There is a high correlation between real earnings and the business cycle. A recession
usually results in a decline in real earnings. However, the impact of the business cycle on
corporate profits varies across industries. Cyclical industries such as automobiles are most
affected, whereas non-cyclical industries such as toothpaste are least affected.
Company specific factors such as the financial structure of the company, quality and
experience of the management, and barriers to entry also matter and determine the impact
of the business cycle on earnings.
Valuation multiples
We use valuation multiples such as P/E and P/B to compare equities across sectors and
analyze the earnings growth prospects of a company.
P/E is the ratio of the current share price to the earnings per share generated by the
company. A high P/E implies that investors are optimistic about the company’s future
growth prospects and are willing to pay a high price. A high P/E could be because of:
 an increase in expectations of future real earnings growth.
 falling real interest rates.
 a fall in inflation expectations.
 a decline in uncertainty about future inflation.
When the economy is doing well P/E ratios tend to be high. Whereas, during a slowdown,
P/E ratios tend to be low.

Commercial real estate
Commercial real estate is priced in the same manner as equity. The pricing equation for
commercial real estate is:
N
s
Pti = ∑ Et [CFt+s
i
]/(1 + lt,s + θt,s + πt,s + γit,s + κit,s + ϕit,s )
s=1
Commercial real estate has bond-like and equity-like characteristics.

 Rental income is like coupon payments on a bond.
 The value of the property at the end of a lease term is uncertain similar to equities.
Another important characteristic of commercial real estate is that they are generally
illiquid.
Commercial real estate is not a very good hedge against bad economic outcomes. Recession
causes real estate prices to fall while growing economic conditions lead to a steep increase
in property prices. Investors will demand a relatively high risk premium because of the
pro-cyclical and illiquid nature of commercial property.

LM06 Analysis of Active Portfolio Management

Measuring value added
Value added is measured through active return. It is the difference between the return on a
managed portfolio and the return of a benchmark.
RA = RP– RB
Another way of expressing active returns is:
N
R A = ∑ Δwi R Ai
i=1
where R Ai = 𝑅𝑖 − 𝑅𝐵 = the active return of each security.
Example: The benchmark consists of 60% in stocks and 40% in bonds, and the active
portfolio consists of 70% in stocks and 30% in bonds. The ex-post returns are 14% for
stocks and 2% for bonds.
To determine the active return, calculate portfolio returns and benchmark return.
R P = 0.7 ∗ 0.14 + 0.3 ∗ 0.02 = 10.4%; R B = 0.6 ∗ 0.14 + 0.4 ∗ 0.02 = 9.2%
Active return = 10.4% − 9.2% = 1.2%
Contribution from stocks = active weight of stocks * active return of stocks = 0.1 ∗
(0.14 − 0.092) = 0.005
Contribution from bonds = active weight of bonds * active return of bonds = −0.1 ∗
(0.02 − 0.092) = 0.007
Choice of benchmark
A benchmark or passive portfolio must have the following three qualities:
 The benchmark is representative of the assets from which the investor will select.
For example, if the investor’s portfolio is composed of large cap US stocks, then an
appropriate benchmark would be S&P 500.
 Positions in the benchmark portfolio can actually be replicated at low cost.
 Benchmark weights are verifiable ex-ante, and return data are timely ex-post. In
other words, the weights of all constituent securities should be known upfront, and
the returns should be known at the end of the period.
Decomposition of value added
A common way of decomposing value added is between value added due to asset allocation
and value added due to security selection.
Let us consider a portfolio of stocks and bonds where the weights allocated to stocks and

bonds differ from the benchmark and each asset class is actively managed by selecting
individual securities. The total value added due to asset allocation and security allocation is
given by:
R A = (Δwstocks R B ,stocks + Δwbonds R B ,bonds ) + (wP ,stocks R A ,stocks + wP ,bonds R A ,bonds )
The first term is the value added from asset allocation and the second term is the value
added from security selection.
Example: The following information is available for a portfolio comprising equities and
bonds:
 Fidelity (equity) mutual fund return: 35.3%
 Benchmark return: 32.3%
 PIMCO (bond) return: -1.9%
 Benchmark return: -2.0%
 Strategic asset allocation: 60% in equities and 40% in bonds.
 Investor’s asset allocation in the actively managed funds: 68% in Fidelity and 32% in
PIMCO.
Given the above information, compute the total value added due to active asset allocation
and security selection.
Portfolio return = 0.68 ∗ 0.353 + (−1.9) ∗ 0.32 = 23.4%
Benchmark return = 0.6 ∗ 0.323 + (−2.0) ∗ 0.4 = 18.6%
Value added = 23.4% - 18.6% = 4.8%
Combined value from asset allocation = 0.08(32.3) + (-0.08)(-2.0) = 2.7%
Combined value added from security selection = 0.68 (3.0) +0.32 (0.1) = 2.1%
Sharpe ratio and Information ratio
Sharpe ratio measures return per unit of risk in absolute terms.

RP– RF
SR P =
STD(R P )
The information ratio is similar to the Sharpe ratio, but it measures the return per unit of
risk in relative terms. It is a measure of active return earned per unit of active risk.
RP– RB RA
IR = =
STD(R P − R B ) STD(R A )
Sharpe ratio as well as the information ratio can be used to measure both ex-ante or
expected, and ex-post or realized active return and risk.
The Sharpe ratio is unaffected by the addition of cash or leverage in a portfolio. However,
the information ratio is affected by the addition of cash or leverage; adding cash to a
portfolio of risky assets will cause the information ratio to decrease. Similarly, adding

leverage will cause the information ratio to increase.

The information ratio of an unconstrained portfolio is unaffected by the aggressiveness of
active weights. For instance, if the active security weights are all multiplied by some
constant c, then the expected active return becomes cR A and the active risk of the altered
portfolio becomes c STD(R A ). The information ratio of an actively managed portfolio will
remain unchanged as can be seen in the equation below:
cR A RA
IR c = = = IR
cSTD(R A ) STD(R A )
Constructing optimal portfolios
An optimal portfolio is constructed by combining the benchmark portfolio with an actively

managed portfolio. This optimal portfolio should have the highest possible Sharpe ratio,
and the Sharpe ratio is calculated as:
SR2P = SR2B + IR2
The optimal amount of active risk is given by:
IR
STD(R A ) = STD(R B ) ∗
SR B
The total risk of the actively managed portfolio is the sum of the benchmark return
variance and active return variance.
STD(R P )2 = STD(R B )2 + STD(R A )2
Example: Consider an actively managed portfolio with an IR of 0.30 and active risk of
8.0%. The benchmark portfolio has a Sharpe ratio of 0.40 and total risk of 16.0%.
1. What is the optimal amount of aggressiveness (active risk) in the actively managed
portfolio?
2. What is the total risk of the portfolio assuming optimal active risk?
Solution:
1. To calculate the active risk, we use this formula:
IR
STD(R A ) = STD(R B ) ∗
SR B
0.3
Active risk = 0.16 ∗ 0.4 = 0.12
2. Total risk = (0.16)2 + (0.12)2 = 0.04

Standard deviation = √0.04 = 20%
Active security returns
The diagram below illustrates the relationship between the three key parameters in the
fundamental law of active management: forecasted active returns (μi ), active portfolio

weights (wi ), and realized active returns (R Ai ).
Interpretation of the diagram:

 The base of the triangle represents the value added through active management.
This is the difference between realized returns on the actively managed portfolio
and the benchmark. If the active weights are zero, that is weights of the individual
assets in the portfolio and the benchmark are equal, then value added is zero. Value
added depends on the correlation coefficient between active weights and realized
active returns.
 Signal quality measures the ex-ante correlation between the forecasted active
returns and realized active returns. This is known as the information coefficient. A
higher IC implies investors have greater ability to forecast returns. The correlation
between N forecasted active returns, μi , and N realized active returns is given by:
R μ
IC = COR ( σAi , σi )
i i
 A portfolio stands to benefit only if the investor’s forecasts are used in the
construction of the portfolio. The transfer coefficient measures the degree to which
the investor’s forecasts are translated into active weights. The risk-weighted
correlation between optimal active weights and forecasted active returns is given
by:
μ
TC = COR (σi , Δwi σi )
i
 Next, we look at the active security weights for uncorrelated active returns, subject
to a limit on the active portfolio risk. The active security weights are given by:

μi σA
Δwi∗ = ∗
σ2i IC√BR
where IC = information coefficient and BR = breadth or the number of independent
forecasts. For instance, if we have a portfolio of four securities and the active
returns of these four stocks are uncorrelated with each other, then the breadth will
be four.
 The forecasted active returns using the Grinold rule are given by:
μi = ICσi Si
where
IC = expected information coefficient
σi = risk of individual security
Si = score for security i. This is the forecast of expected returns. A positive score
represents the stock is expected to outperform while a negative score indicates the
stock is forecasted to underperform. If there are four stocks with scores of +2, +1, -1
and -2, then it implies that the stock with a score of +2 will outperform the most.
The full fundamental law
According to the full fundamental law, the expected active return is expressed as:
E(R A ) = (TC)(IC)√BR σA
Information ratio is expressed as:
IR = (TC)(IC)√BR.
The transfer coefficient is calculated as:
TC = COR(μi /σi , Δwi σi )
It can take values from -1 to +1.
If there are no constraints and the actual portfolio optimal weights are equal to the actual
weights, then TC will be equal to 1 and we will have the basic fundamental law.
E(R A )∗ = (IC)√BR σA and IR = (IC)√BR
The optimal amount of active risk in an actively managed portfolio with constraints is
expressed as:
IR∗
σA = TC ∗ ∗ σB
SR B
The maximum value of the constrained portfolio’s Sharpe ratio is given as:
SR2P = SR2B + (TC)2 (IR∗ )2
Example: Consider an actively managed portfolio has a transfer coefficient of 0.50 and an
unconstrained information ratio of 0.30. The benchmark portfolio has a Sharpe ratio of
0.40 and risk of 16.0%. What is the optimal amount of aggressiveness in the actively

managed portfolio?
Solution:
IR∗
Optimal amount of active risk σA = TC ∗ ∗ σB
SRB
0.3
σA = 0.5 ∗ 0.4 ∗ 16% = 6%
If the actively managed portfolio is constructed with this amount of active risk, what is the
Sharpe ratio?
Solution:
SR2P = SR2B + (TC)2 (IR∗ )2
SR2P = 0.42 + 0.52 ∗ 0.32 = 0.1825
SR P = 0.43
If the constrained portfolio has an active risk of 8.0%, how can the active risk be lowered to
the optimal level of 6.0%?
Solution:
The benchmark has a risk of 0% and the constrained portfolio has an active risk of 8.0%. To
6.0%
get an optimal level of 6.05, the weight of the actively managed fund must be 8.0% = 75%.
The actively managed portfolio will have an optimal risk of 6.0% if the weight in the
benchmark is 25% and 75% in the actively managed fund.
Ex-post performance measurement
Actual performance is measured by the relationship between relative weights and realized
active returns.
Expected value added conditional on the realized information coefficient, ICR , is given by:
E(R A |ICR ) = (TC)(ICR )√BR σA
The realized value added of an actively managed portfolio can be divided into two parts:
1. The expected value added given the realized skill of the investor that period.
2. Any noise that results from constraints that affect the optimal portfolio structure.
R A = E(R A |ICR ) + Noise
The equation also implies that a portfolio’s ex-post active return can be decomposed into:
1. Variation due to the realized information coefficient: E(R A |ICR )
2. Variation due to a constraint-induced noise
Applications of the fundamental law
The fundamental law has been used in time-series and cross-sectional applications.
 Selection of country equity markets in a global equity fund is an example of a cross-

sectional application of the fundamental law.

 The timing of credit and duration exposures in a fixed-income fund is an example of
time-series and cross-sectional application of the fundamental law.
For an unconstrained portfolio, information ratio does not change with active risk as active
return changes proportionately with active risk.
For a constrained portfolio, IR generally decreases as the active risk increases. As active
risk increases, TC decreases, which causes IR to decrease.
Practical limitations
The fundamental law ignores transaction costs and taxes. The fundamental law is based on
MVO and hence has many of the same limitations apply. This section focuses on two
limitations: 1) IC as an ex-ante measure of skill and 2) independence of investment
decisions.
Some limitations of using IC as an ex ante measurement of skill are:
 Investors overestimate their own skill.
 Forecasting ability varies across different asset segments.
 Forecasting ability varies over time.
 Security selection strategies have produced results which are 45% - 91% of original
estimates using the fundamental law.
 If uncertainty about forecasting ability is high the expected value added will be low.
Some limitations due to lack of independence in investment decisions are:
 N is not an adequate measure of BR when active returns of individual assets are
correlated, for example:
o Decisions to overweight all stocks in a given industry.
o Bond returns tend to be highly correlated.
 With hedging strategies BR can be larger than N.

LM07 Trading Costs and Electronic Markets

Components of execution costs
Transaction costs include explicit costs and implicit costs.
‘Explicit costs’ are the direct costs of trading. They include broker commissions,
transaction taxes, stamp duties, and exchange fees.
‘Implicit costs’ are indirect costs caused by the market impact of trading. They result from
the following issues:
 Bid-ask spread
 Market impact
 Delay costs (also called slippage)
 Opportunity costs (or unrealized profit/loss)
Effective spreads
A dealer’s bid-ask spread is calculated as ask price minus bid price. If several dealers offer
quotes in a market:
 The best bid is the highest bid price. It is also called the inside bid.
 The best ask is the lowest ask price. It is also called the inside ask.
 The spread between the best bid price and the best ask price is the market bid-ask
spread. It is also called the inside spread.
 The midquote price is the average of the market bid and ask prices.
The effective spread is two times the difference between the trade price and the midquote
price before the trade occurred.
 It is a good estimate of actual transaction costs when orders are filled in single
trades.
 It is a poor estimate of actual transaction costs when large orders have been split
into many parts to fill over time or when small orders receive price improvement.
Implementation shortfall
The implementation shortfall method compares the values of the actual portfolio with that
of a paper portfolio constructed on the assumption that trades could take place at the
prices that prevailed when the decision to trade was made. The excess of the paper value
over the actual value is the implementation shortfall.
It measures the total cost of implementing an investment decision by capturing all explicit
and implicit costs. So, market impact costs, delay costs, as well as opportunity costs which
are not captured in the effective spread method are captured by the implementation
shortfall method.

VWAP transaction cost estimates

The volume-weighted average price (VWAP) method of estimating transaction costs
compares average fill prices to average market prices during a period surrounding the
trade.
 The advantage of VWAP over implementation shortfall is that it is easy to interpret.
It answers the question: ‘Did you get a better or worse average price than all traders
trading when you were trading?’
 The disadvantage of VWAP is that it will be close to zero if trades being evaluated
represent a large fraction of the trading volume or if trades took place at the same
rate as other trades in the market.
VWAP generally gives a lower transaction cost estimate than implementation shortfall.
Therefore, it is more popular with investment managers as it helps them convey that
transaction costs are low.
Advantages of electronic trading systems

The advantages of electronic order-matching systems over floor-based trading systems
include:
 Cost and operational efficiency
 Lack of human bias
 Keep perfect audit trials
 Keep hidden orders perfectly hidden
 Can operate on a continuous basis
 Can operate when bad weather or other events would prevent workers from
operating the floor
Additionally, computers now dominate many trading strategies because they:
 Have an infinite attention span and a very wide attention scope
 Can respond very quickly
 Are perfectly disciplined and do as told (programmed)
 Do not forget
Electronic markets for stocks, futures, and options are quite common. However, the bond
market has not received much attention and most bond investors still trade with dealers in
OTC markets.
Market fragmentation
Market fragmentation occurs when the same instrument is traded across multiple venues.
Markets across the world have become increasingly fragmented. Market fragmentation
increases the potential for price and liquidity discrepancies.
Electronic algorithmic trading techniques, such as liquidity aggregation and smart order

routing, help traders manage the challenges and opportunities presented by fragmented
markets.
 Liquidity aggregation involves creating ‘super books’ that present liquidity across
all markets for a given instrument. This helps traders identify the best
opportunities.
 Then smart order-routing algorithms are used to send orders to the markets that
display the best-quoted prices and sizes.
Types of electronic traders

The major types of electronic traders include:
 Electronic news traders: These traders profit based on news. They subscribe to
high speed electronic news services. They process news releases extremely fast and
place trade based on their analysis.
 Electronic dealers: Like other dealers, electronic dealers also earn profits through
their bid-ask spreads. However, at the first sign that prices may move against their
inventory position, they immediately take liquidity by executing on the opposite
side and reduce their exposure.
 Electronic arbitrageurs: These traders monitor multiple markets to look for
arbitrage opportunities.
 Electronic front runners: These are low-latency traders who use AI methods to
identify when large traders, or many small traders, are trying to fill orders on the
same side of the market. If they anticipate a lot of buy orders coming, they will
purchase the security before these buy orders are placed by the other market
participants.
 Electronic quote matchers: These traders exploit the option value of standing
orders. Standing orders are limit orders waiting to be filled. Quote matchers buy
when they believe they can rely on standing buy orders to get out of their positions
quickly.
Characteristics and uses of electronic trading systems

To trade effectively, electronic traders do not simply need to be fast, but they have to be
faster than their competitors.
Electronic traders use various strategies to minimize their communication times. These
strategies involve minimizing communication distances and maximizing line speeds. To
minimize communication distances, electronic traders locate their computers as close as
possible to the exchanges. Collocation is a practice where exchanges allow electronic
traders to place their servers in the rooms where the exchange server operates. Exchanges
charge substantial fees for collocation service.
Electronic traders minimize their computational latencies by using the following strategies

 Using very fast computers

 Creating contingency tables that contain a prearranged action plan
 Using very efficient software
Traders often use electronic brokers for advanced orders, trading tactics and algorithms.
Some commonly used strategies are: hidden orders, quote leapfrogging, flickering quotes,
and machine learning.
Low-latency traders
Latency is the elapsed time between the occurrence of an event and a subsequent action
that depends on that event. Latency depends on three steps:
1. Communication in: Recognizing that the event took place.
2. Computation: Processing the information.
3. Communication out: Responding and acting on the information.
To reduce the total latency, low-latency traders try to minimize their latency at every step
by using very fast communication and computation systems.
Risks associated with electronic trading

Electronic trading has largely benefited investors through higher trade processing
efficiencies and lower transaction costs. However, it has also created new systemic risks
that are a cause for concern for regulators and practitioners.
Some examples of systemic risks caused by electronic trading include:
 Runaway algorithms produce streams of unintended orders caused by
programming mistakes.
 Fat finger errors occur when a manual trader submits a larger order than intended.
 Overlarge orders demand more liquidity than the market can provide.
 Malevolent order streams are created deliberately to disrupt the markets.
To mitigate these systemic risks, the following strategies can be used:
 Thorough testing of an algorithm before it is deployed.
 Rigorous market access control must ensure only orders coming from approved
sources enter the order matching systems.
 Ensure that an order meets pre-set criteria before it is entered on an exchange.
 Brokers should surveil all client orders to ensure that they are reasonable.
 Some exchanges have adopted price limits and trade halts to stop trading in highly
volatile conditions.
Real-time surveillance for abusive trading practices

Real-time surveillance of markets often can detect order front running and various market
manipulation strategies.

Market manipulators use improper activities such as the following:

 Trading for market impact involves trading to raise or lower prices deliberately.
 Rumormongering is the dissemination of false information to alter other investors’
value assessments.
 Wash trading consists of trades arranged among commonly controlled accounts to
create the impression of market activity at a particular price.
 Spoofing, also known as layering, is a trading practice in which traders place
exposed standing limit orders to convey an impression to other traders that the
market is more liquid than it is or to suggest to other traders that the security is
under- or overvalued.
Market manipulation strategies include:
 Bluffing: involves submitting orders and arranging trades to influence other traders’
perceptions of value. For example, pump-and-dump
 Gunning the market: a strategy used by market manipulators to force traders to do
disadvantageous trades. For example, selling quickly to push prices down with the
hope of triggering stop-loss sell orders.
 Squeezing and cornering: manipulator obtains control over resources necessary to
settle trading contracts. The manipulator then unexpectedly withdraws those
resources from the market, which causes traders to default on their contracts, some
of which the manipulator may hold. The manipulator profits by providing the
resources at high prices or by closing the contracts at exceptionally high prices. For
example, short squeeze.

Ethics 2023 Level II High Yield Notes
LM01 Code of Ethics and Standards of Professional Conduct

CFA Institute Professional Conduct Program
Structure
All CFA Institute members and candidates must comply with the Code and Standards.
The CFA Institute Board of Governors maintains oversight and responsibility for the
Professional Conduct Program (PCP). The Disciplinary Review Committee (DRC) is
responsible for enforcement of Code and Standards.
 DRC is a volunteer committee of CFA charterholders.
 DRC partners with Professional Conduct staff to establish and review professional
conduct policies.
 DRC is also responsible for reviewing conduct when there is a potential violation.
The CFA Institute Bylaws and Rules of Procedure for Professional Conduct (Rules of
Procedure) form the basic structure for enforcing the Code and Standards.
Process for Enforcement
Professional Conduct inquiries can come from:
 Self-disclosures by members/candidates on their annual Professional Conduct
Statements.
 Written complaints received by the Professional Conduct staff.
 Evidence of misconduct received through public sources such as media or
regulatory notices.
 A report by a CFA exam proctor.
 Analyzing exam material and monitoring social media.
Once an inquiry is initiated, the Professional Conduct staff may:
 Request a written explanation from the member/candidate.
 Interview the member/candidate, the complainant or other third parties.
 Collect documents and records relevant to the investigation.
After investigating, the Professional Conduct staff may:
 Take no disciplinary action.
 Issue a cautionary letter.
 Propose a disciplinary sanction.
If a disciplinary sanction is proposed, the member/candidate can either accept or reject the
sanction. If rejected, the matter is referred to a panel of Disciplinary Review Committee
members for a hearing. The panel then determines if a violation occurred and if so, what
sanction should be imposed.

Sanctions imposed may include:

 Public condemnation.
 Suspension of membership and revocation of CFA charter (for members).
 Prohibition from further participation in the CFA Program (for candidates).
Six Codes of Ethics and Seven Standards of Professional Conduct

Six Codes
Members/candidates must:
 Act with integrity, competence, diligence, and respect and in an ethical manner with
the public, clients, prospective clients, employers, employees, colleagues in the
investment profession, and other participants in the global capital markets.
 Place the integrity of the investment profession and the interests of clients above
their own personal interests.
 Use reasonable care and exercise independent professional judgment when
conducting investment analysis, making investment recommendations, taking
investment actions, and engaging in other professional activities.
 Practice and encourage others to practice in a professional and ethical manner that
will reflect credit on themselves and the profession.
 Promote the integrity and viability of the global capital markets for the ultimate
benefit of society.
 Maintain and improve their professional competence and strive to maintain and
improve the competence of other investment professionals.
Seven Standards
I. Professionalism
II. Integrity of Capital Markets
III. Duties to Clients
IV. Duties to Employers
V. Investment Analysis, Recommendations, and Actions
VI. Conflict of Interest
VII. Responsibility as a CFA Institute Member or CFA Candidate

LM02 Guidance for Standards I-VII

Ethical responsibilities required by the Code and Standards
I. Professionalism
A. Knowledge of the law
 Understand and comply with all applicable laws, rules and regulation.
 In a case of a conflict, comply with the stricter law.
 Do not knowingly participate in any violation. Disassociate from such activity.
B. Independence and objectivity
 Use reasonable care and judgment.
 Maintain independence and objectivity.
 Do not offer, solicit or accept gifts; however, small token gifts are ok.
C. Misrepresentation
 Do not guarantee investment performance.
 Avoid plagiarism (the practice of taking someone else’s work or ideas and
passing them off as one’s own).
 Do not omit important facts.
D. Misconduct
 Do not lie, cheat, steal or behave in a manner that affects your professional
reputation or integrity.
II. Integrity of capital markets
A. Material nonpublic information
 Do not act or help others to act on material nonpublic information. (Information
which would be likely to affect a stock's price once it becomes known to the
public.)
 However, mosaic theory (material public information + nonmaterial nonpublic
information) is not a violation.
B. Market manipulation
 Do not manipulate prices/trading volumes to mislead other market participants.
 Do not spread false rumors.
III. Duties to clients
A. Loyalty, prudence, and care
 Act with reasonable care and exercise prudent judgment.
 Place the client’s interest before your employer’s or your interests.
 Soft dollars must be used for the benefit of the client.

 Seek the best execution.

 Vote proxies in the best interest of clients.
B. Fair dealing
 Do not discriminate against any clients when disseminating recommendations
and taking investment action.
 Different level of service is allowed, as long as it does not negatively affect any
client.
 Different service levels should be disclosed to all clients and prospects.
C. Suitability
 In advisory relationships, develop and update an IPS periodically. Understand
the client’s risk profile.
 In fund/index management, ensure that investments are consistent with the
stated mandate.
D. Performance presentation
 Do not misstate performance.
 Make detailed information available on request.
E. Preservation of confidentiality
 Maintain confidentiality of current, former and prospective clients.
 Unless (1) disclosure is required by law (2) information concerns illegal
activities by a client (3) client permits the disclosure.
IV. Duties to employers
A. Loyalty
 Do not harm your employer.
 Obtain written consent from the employer before starting an independent
practice.
 Do not take confidential information, client lists, financial models, etc. when
leaving an employer.
B. Additional compensation arrangements
 Do not accept gifts, benefits or compensation that will create a conflict of interest
with your employer.
 You may accept them if you obtain written consent from all parties involved.
C. Responsibilities of supervisors
 Prevent employees under your supervision from violating applicable laws, rules,
regulations and the Codes and Standards.
V. Investment analysis, recommendations, and actions

A. Diligence and reasonable basis

 Have a reasonable and adequate basis for any investment analysis,
recommendation or action (even when using third party research).
B. Communication with clients and prospective clients
 Tell clients about your investment process.
 Distinguish between fact and opinion.
C. Record retention
 Maintain records (Standards recommend storing records for at least 7 years).
VI. Conflicts of interest
A. Disclosure of conflicts
 Disclose conflict of interest in plain language.
B. Priority of transactions
 Client transactions come before employer transactions which come before
personal transactions.
 Avoid front running.
 Fee-paying family member should be treated no different than any other client.
C. Referral fees
 Disclose referral arrangements to clients, prospective clients and employers.
 Disclosure of referral fees helps the clients evaluate any possible partiality
shown in the recommendation of service.
VII. Responsibilities as a CFA Institute member or CFA candidate
A. Conduct as participants in CFA Institute programs
 Don’t cheat on the exams.
 Keep questions and exam information confidential.
B. Reference to CFA Institute, the CFA designation, and the CFA program
 Fill the professional conduct statement and pay membership dues annually.
 References to partial designation not allowed (wrong usage: I am a CFA Level I).
However, you can say that you have passed Level I, II or III.
 Not to be used as a noun. Only use it as an adjective.
 Do not state that holders of CFA charter are better than others or that they
produce better investment results.

LM03 Application of the Code and Standards

This reading has seven cases that demonstrate how to apply the Code and Standards in
situations where professional and ethical judgment is required. In this summary, we have
covered the most important points of these cases.
Serengeti Advisory Services

Serengeti Advisory Services (Serengeti) is a Tanzania-based equity research firm. It was
founded five years ago by three CFA charterholders: Fatima Bashar, Hasini Shah, and
Amara Kariuki. The firm conducts investment research on listed African companies and
sells the research to institutional asset managers on an annual subscription basis. Each
month, subscribers receive two or three research reports. Bashar, as CEO and head of
research, has recently launched a premium subscription service. Clients can receive six
additional research reports of their choice per year for an additional fee. Because of the
cost of the premium service, Bashar offers it only to clients that she believes can afford it.
(Question 1)
Serengeti’s research analysts cover small-to mid-cap African companies. For companies not
covered by Serengeti, Bashar has created an “approved list” of broker/dealers from across
the continent whose research she believes meets Serengeti’s standards. Bashar informs her
client that she has put in place the following procedures to ensure that third-party research
is used properly:
 Procedure 1: When a research report, ours or theirs, cites specific quotations as
attributable to “leading analysts” and “leading experts,” I require the analyst who wrote
the report to name the specific references.
 Procedure 2: When a research report contains statistical estimates of forecasts
prepared by others (such as economists) and identifies the source, I require the analyst
who wrote the report to remove any qualifying statements or caveats that may have
been used.
 Procedure 3: When a research report contains charts and graphs prepared by others, I
require the analyst who wrote the report to cite their sources. (Question 2)
To be on Serengeti’s approved list, broker/dealers must undergo a rigorous due diligence
process carried out by Bashar’s team and comply with Procedures 1–3. Bashar has
established the following referral arrangement with the firms on the approved list:
Whenever Serengeti’s analysts prepare a research report using information sourced from
an approved broker/dealer, Bashar encourages the firm’s clients to trade through that
broker/dealer. Bashar believes this referral arrangement, which she discloses to clients
and prospective clients, provides additional incentive for broker/dealers to share insights
and information with her analysts at no cost. Broker/dealers seek to be added to
Serengeti's "approved list" because of the firm's large client base. (Question 3)
Bashar has the following phone conversation with Jon Grant, CFA, of Grant Asset

Management, one of Serengeti’s largest clients.

Grant: I have found the research partnerships that Serengeti has established with
broker/dealers to be very beneficial. The due diligence that you conduct on the
broker/dealers on your approved list saves me from investigating these firms on my own.
As a result, I always follow the analysts’ investment recommendations and execute my
clients’ trades through the broker/dealer that produced the research report or whose
information was used in your research. I noticed that Olatunji Financial, one of the largest
broker/dealers in Nigeria, is not on your approved list. Why is that?
Bashar: Thank you so much for the compliment and your support. I am glad that Serengeti
can provide this service, and I know the broker/dealers also appreciate your business. I
have heard that Olatunji Financial conducts outstanding research on Nigerian-listed
companies, but we have not had an opportunity to work with them. (Question 4)
A few days later, Bashar receives an email from Amope Olatunji, CFA, president and CEO of
Olatunji Financial Services (OFS). In the email, Olatunji states, “Jon Grant has told me about
the great equity research that your firm produces. I would like OFS to be added to
Serengeti’s approved list. To this end, OFS would like to invite you and two of your
colleagues to our upcoming West Africa Investment Forum that we host at our
headquarters in Lagos, Nigeria. We will pay all of your expenses related to attending the
forum, including air travel and overnight hotel accommodations.”
Bashar selects Kariuki, the head of business development and Shah, the chief compliance
officer for the event. During a meeting to discuss how to proceed with Olatunji’s invitation,
Bashar states, “As I see it, there are three possible options we have:
Option 1 We allow OFS to pay for all of our expenses.
Option 2 We allow OFS to pay for our air travel and hotel accommodations, and we pay for
our meals and entertainment.
Option 3 We allow OFS to pay for the forum’s meals and entertainment, and we pay for our
air travel and hotel accommodations.” (Question 5)
At the forum, Bashar, Shah, and Kariuki separate to talk with the attendees. When Kariuki
walks over to the recycling bin to dispose of her water bottle, she notices a document on
top of the bin. When she picks up the document, she realizes it is a list of OFS clients,
complete with contact information, email addresses, and average trading volumes over the
previous three years. After taking a photo of the document with her phone, she places it
back in the bin and proceeds to seek out those clients with the highest trading volumes.
(Question 6)
The next morning, Bashar, Shah, and Olatunji meet to discuss the possibility of OFS being
added to Serengeti’s approved list. The following are excerpts from their meeting:
Olatunji: Bashar, I propose you open a discretionary personal account with OFS. I will use

the cash that you deposit into the account to purchase a “model portfolio” based on the
Nigerian equities our analysts are recommending. Any time our analysts change their
recommendations or issue new ones, I will execute trades in your account when we do it
for our other discretionary clients. I will also send you an instant message before I make
the trade just to let you know, and then I will call you afterward to discuss the trade in
more detail. With this arrangement, I can prove to you how well our analysts’
recommendations perform, and you will be able to participate directly.
Bashar: That’s a good idea. Could you send me portfolio performance data monthly as you
do with your other clients? Please be aware that I will not disclose this model portfolio to
Serengeti’s clients until I see how well your analysts’ recommendations perform. In the
meantime, Shah will continue to conduct due diligence on OFS to determine whether it
should be added to our approved list of broker/dealers. (Question 7).
Bashar opens a discretionary personal account with OFS the next day. Despite market
liquidity issues, Olatunji is able to build the model portfolio in Bashar's account within a
few days. Soon after, Olatunji calls Bashar.
Olatunji: I have good news for you. At the end of next week, our team of analysts will be
issuing a strong buy recommendation on Swann Bank (SWNB), a commercial bank that is
headquartered in Lagos. Because its stock is thinly traded, I have already started building a
position in SWNB in your portfolio. Three days before the research report is issued, I have
instructed our traders to simultaneously buy and sell shares of SWNB to increase its
liquidity in the market so it will be easier for our clients to purchase it for their portfolios.
(Questions 8, 9, 10)
Bashar: Thank you for the information and update.
Later that day, after Olatunji has established a position in SWNB in Bashar's portfolio,
Bashar contacts Grant to inform him of Olatunji's upcoming recommendation. (Question
11)
Case Questions
1. Does Bashar violate the CFA Institute Code and Standards with regard to the
premium subscription service?
Yes, Bashar violates Standard III(B): Fair Dealing because she offers the premium
subscription service only to clients who she believes can afford it. The service should be
offered to all clients and prospective clients, it should not be offered selectively.
2. Which of Bashar’s procedures regarding the use of third-party research violates
the CFA Institute Code and Standards?
Procedure 2 violates Standard I(C): Misrepresentation by asking the analyst who wrote the
report to remove qualifying statements or caveats from reports. According to Standard
I(C), when presenting statistical estimates of forecasts prepared by others, the source

should be identified along with the qualifying statements or caveats that may have been
included.
3. Does Bashar’s referral arrangement with the broker/dealers on the approved list
violate the CFA Institute Code and Standards?
No. Bashar is merely suggesting to the firm’s clients that they trade through a particular
broker/dealer because of the research or other information that the firm provided. The
asset manager client makes the final decision on where to direct its trades.
4. With regard to Grant and Bashar’s phone conversation about Serengeti’s research
partnerships, who violated the CFA Institute Code and Standards?
Only Grant violated the CFA Institute Code and Standards. Grant violated Standard III(A):
Loyalty, Prudence, and Care by solely relying on Bashar’s suggestions about which
broker/dealer to use to execute his clients’ trades instead of seeking the broker/dealer that
provides best price and best execution.
5. To avoid violating the CFA Institute Code and Standards, which of Bashar’s
responses to Olatunji’s forum invitation would be most appropriate?
Option 3. To prevent violating Standard I(B): Independence and Objectivity, Serengeti
should pay for air travel and overnight hotel accommodations. Under Standard I(B), it
would be permissible for OFS to pay for meals and entertainment provided at the forum.
This action is unlikely to jeopardise their independence and objectivity in deciding whether
to add OFS to Serengeti's approved list of broker/dealers.
6. Did Kariuki violate the CFA Institute Code and Standards by seeking out OFS
clients on the list with the highest trading volumes at the forum?
Yes. Kariuki violated Standard I(D): Misconduct by using OFS’s confidential and proprietary
information that she found on top of the recycling bin.
7. With respect to the creation and management of the model portfolio, have Bashar
or Olatunji most likely violated the CFA Institute Code and Standards?
Neither Bashar nor Olatunji violated the CFA Institute Code and Standards.
8. In telling Bashar about the upcoming recommendation on SWNB, is Olatunji
violating the CFA Institute Code and Standards?
Yes. Olatunji violated Standard III(B): Fair Dealing by telling Bashar about the “strong buy”
recommendation on SWNB before the research report is disseminated to all clients.
9. In building a position in SWNB in Bashar’s portfolio, have Olatunji and Bashar
violated the CFA Institute Code and Standards?
Yes. Both Olatunji and Bashar violated the CFA Institute Code and Standards.
Olatunji violated Standard III(B): Fair Dealing by purchasing shares of SWNB in Bashar’s

model portfolio before the recommendation is disseminated to other clients.

Bashar violated Standard I(A): Knowledge of the Law. Bashar knows that Olatunji is
building a position in SWNB in her portfolio before the research report on SWNB is issued.
She should have rejected the advance purchase of SWNB.
10. In his instructions to his traders, Olatunji most likely violated all of the following
CFA Institute Standards of Professional Conduct except:
A. Market Manipulation.
B. Priority of Transactions.
C. Responsibilities of Supervisors.
B is correct.
By instructing his traders to simultaneously buy and sell shares of SWNB (called “wash
trading,” which creates misleading and artificial activity in the stock) just before the
research report is issued, Olatunji violated Standard II(B): Market Manipulation, as well as
Standard IV(C): Responsibilities of Supervisors
Olatunji did not violate Standard VI(B): Priority of Transactions. In this scenario, Olatunji
did not trade for himself, beneficially or otherwise.
11. By informing Grant about the recommendation on SWNB, Bashar most likely
violated all of the following CFA Institute Standards of Professional Conduct except:
A. Fair Dealing.
B. Priority of Transactions.
C. Preservation of Confidentiality.
C is correct.
Bashar violated Standard III(B): Fair Dealing by telling Grant about the strong buy
recommendation before telling her other clients.
By waiting until the position in SWNB is established in her portfolio before telling Grant
about the upcoming recommendation, Bashar violated Standard VI(B): Priority of
Transactions.
Bashar did not violate Standard III(E): Preservation of Confidentiality, which addresses
keeping information about current, former, and prospective clients confidential.
Banco Libertad
Banco Libertad (BL) is a private bank based in the country of Urutina. It provides
investment management, securities research, real estate financing, and wealth
management services to high-net-worth individuals.
Sofia Maduro, CFA, is the managing director of BL’s investment management division. This
morning, Maduro is meeting with two of the portfolio managers that she supervises: Julio

Ortiz, CFA, and Guadalupe Sanchez, CFA.

All of Ortiz’s clients reside in either Urutina or Chiladour, and all of Sanchez’s clients reside
in either Urutina or Panaguay. For compliance purposes, applicable law for them is the
country where their clients reside. Panaguay and Chiladour are neighboring countries.
Panaguay has nascent capital markets and Chiladour is a politically unstable country. As a
result, wealthy clients from these countries prefer to invest their money in Uruntina.
Exhibit 3 contains a comparison of the securities regulations and laws in each country as
well as BL’s policies.
The following conversation takes place at the meeting:

Maduro: Thank you for taking the time out of your busy schedules to meet with me. It must
be difficult to manage the portfolios of clients who live in different countries and in
different time zones. I want to commend both of you on your use of technology to
communicate with your clients. Your use of social media to post investment
recommendations, performance reports, and investment opinions has set a great example
for other managers at BL.
Ortiz: Thank you for the compliment and for allowing Guadalupe and I to use instant
messaging (IM) platforms to manage our clients’ portfolios. Now, all our clients have to do
is IM us whenever they want to buy and sell securities, and we, in turn, IM them with their

trade confirmations once the transactions are complete.

Maduro: On a more serious note, it has been brought to my attention that law enforcement
officials from the governments of Chiladour and Panaguay have contacted both of you
requesting confidential information on your clients. What are you doing about these
inquiries?
Ortiz: I have a client who is the former Minister of Finance in Chiladour. Since establishing
his own financial consulting business five years ago, he has deposited more than USD5
million with BL. Now he is under investigation for allegedly embezzling from the Chiladour
Treasury. In response to the inquiry into his personal finances, I have given law
enforcement officials from Chiladour all of the information they requested on this client.
Sanchez: I have a client who is a general in the Panaguay Army. His grandfather founded
the country’s second-largest petroleum processing facility. When the general opened his
account last summer, he deposited approximately USD35 million and told me that he had
inherited the money from his grandfather, who had recently passed away. According to law
enforcement officials in Panaguay, this money allegedly came from bribes the general
received from directing military contracts to specific corporations. Like Ortiz, I was so
disgusted by the client’s behavior that I gave the law enforcement officials all of the
information they requested on this client. (Questions 1, 2, 3)
Maduro: Thank you for doing this. It is important that BL maintain good relations with law
enforcement officials throughout the region. Speaking of good relations, I noticed that
many of your clients have been referred by law firms in Chiladour and Panaguay. Do you
have any referral arrangements with these firms?
Ortiz and Sanchez: Our referral arrangements are very informal, which is why we do not
bother disclosing them to clients. Every year we purchase tickets for the principals of these
firms to attend CONMEBOL COPA AMERICA, the men’s international football tournament.
In addition, we always take them out to dinner whenever we travel to Chiladour and
Panaguay to meet with our clients or when the principals come to Urutina to meet with
their clients. (Question 4)
Maduro: Do you have any other arrangements that I am not aware of?
Ortiz and Sanchez: Now that you mention it, yes we do. We have special arrangements with
two of our largest clients, Juan Fabre and Caroline Zeissl. After our securities analysts post
changes to their research recommendations on the BL client website, we immediately call
these two to discuss these changes in detail and any impact on their portfolios. Clients can
sign up for email or text alerts to notify them when analyst recommendations change, but
we feel calling Fabre and Zeissl personally is important because of their asset size. In
addition, whenever Fabre and Zeissl are interested in participating in an IPO, we invite
them to the company roadshow presentations organized by the investment bank
underwriting the issue. Because of capacity constraints, we cannot invite our other clients

who are interested in these IPOs to the presentations, so we do not disclose the roadshow
opportunity to them. Whether or not they receive an invitation, we make IPO opportunities
available to all our clients for whom the investment is suitable. (Questions 5, 6)
Ortiz: I have known and worked with Fabre and his family for about 15 years. In any year
that his portfolio outperforms its benchmark, he allows my family and I to use his beach
house in Jamaica for a week of our choosing. This arrangement is acknowledged in writing
by the client and documented as part of the client’s investment policy statement (IPS).
Three months ago, my family and I used his beach house because, for the first time since he
made the offer, his portfolio outperformed its benchmark this past year.
Sanchez: I have known and worked with Zeissl and her family for about 10 years. The Zeissl
family owns the Aneka, a luxury hotel and spa in the Bahamas. She has said that in any year
that her portfolio outperforms its benchmark, my family and I can spend a week at the
Aneka without charge. As Julio described, this arrangement is documented with the client.
Her portfolio also outperformed its benchmark this past year, so my family and I spent a
week at the Aneka this past summer.
Maduro: Not having reviewed your client files, I was unaware of these arrangements, so
thank you for telling me about them. I do have another meeting shortly, so let’s end for
today. (Question 7)
Case Questions
1. Have Ortiz and/or Sanchez violated the CFA Institute Code and Standards in their
use of social media?
Yes. Sanchez has violated Standard I(A): Professionalism–Knowledge of the Law because
the use of social media to post investment information is prohibited in Panaguay, where
applicable law is stricter than the Code and Standards.
Ortiz has not violated the Code and Standards because the use of social media to post
investment information is allowed in Chiladour and Urutina.
2. Has Maduro violated the CFA Institute Code and Standards by allowing Ortiz and
Sanchez to use social media to post investment information?
Yes, Maduro has violated Standard IV(C): Duties to Employers–Responsibilities of
Supervisors. Maduro has allowed, and in doing so approved of, Sanchez’s use of social
media with her clients in Panaguay, where it is prohibited for posting investment
information (i.e., applicable law is stricter than the Code and Standards).
3. Have Ortiz and Sanchez violated the CFA Institute Code and Standards by using
instant messaging and disclosing information on their clients to law enforcement
officials?
Yes. Ortiz and Sanchez violated the Code and Standards by using IM to communicate with
their clients and by disclosing information on their clients to law enforcement officials.

Both are violations of applicable law and therefore violations of the Code and Standards.
4. Have Ortiz and Sanchez violated the CFA Institute Code and Standards in their
referral arrangement with lawyers in Chiladour and Panaguay?
Yes. Ortiz and Sanchez have violated Standard VI(C): Conflicts of Interest–Referral Fees by
not disclosing to BL, clients, and prospective clients the benefit they receive and pay to the
lawyers who recommend clients to them.
5. Do Ortiz and Sanchez violate the CFA Institute Code and Standards by calling Fabre
and Zeissl to discuss changes in research recommendations made by BL’s analysts?
No. Ortiz and Sanchez have not violated the Code and Standards. They call Fabre and Zeissl
after the changes in recommendations have been posted on BL’s website (i.e., publicly
posted or made publicly accessible). Therefore, they are not treating their other clients
unfairly but rather are calling their largest clients to discuss the changes in more detail.
6. Have Ortiz and Sanchez violated the CFA Institute Code and Standards by inviting
Fabre and Zeissl to IPO roadshow presentations?
No. Ortiz and Sanchez have not violated the Code and Standards. Inviting clients to the IPO
roadshow is not a violation of the Code and Standards. Fabre and Zeissl are their largest
clients, and in addition, being invited to the IPO roadshow does not disadvantage other
clients who are interested in the IPO.
7. Have Ortiz and Sanchez violated the CFA Institute Code and Standards in not
disclosing their respective arrangements with Fabre and Zeissl?
Yes. Ortiz and Sanchez violated Standard IV(B): Duties to Employers–Additional
Compensation Arrangements. Although they obtained written acknowledgement from
Fabre and Zeissl at the time each client made their offer, and included it as part of the IPS,
they did not obtain their employer’s written consent for the arrangement.
QuantHouse
Case facts
Daniel Singh, PhD, CFA works at QuantHouse (QH) a global investment firm that uses
quantitative techniques to implement investment strategies. Singh created the Artificial
Trading Model (ATM) to make investment decisions largely without human interaction.
The ATM has three components: an Alpha Model, a Risk Model, and an Optimizer.
Singh uses the ATM exclusively for QH’s institutional clients and does not mention it when
talking with his high-net-worth individual clients. Singh and his team update the Alpha and
Risk components of the ATM on a quarterly basis and the Optimizer on an annual basis.
Some of QH’s institutional clients are not happy with the ATM. In response to these
complaints, QH’s director of research and Singh’s supervisor, Charlotte Ringfield, CFA, asks
Singh to review the model. Upon review, Singh and his team find that some of the Risk

Model components are sending information to the Optimizer in decimal form while other
components are sending information in percentages. This improper scaling has resulted in
the Optimizer giving inappropriate weights to some of the common risk factors. Singh
presents his findings to Ringfield and advices that the error be fixed as soon as possible.
But Ringfield disagrees and tells him to correct the error when the Risk Model is updated at
the end of the quarter. She also asks Singh to temporarily disable the common risk factors
in the Risk Model until the model is updated. Until then, she asks Singh and his team not to
mention the error to others.
Two weeks later, after a very turbulent period in the financial markets, more clients
complain about their portfolios’ underperformance. When clients inquire about their
portfolio’s performance, Ringfield attributes the performance to market volatility and the
functioning of the model’s common risk factors.
Disturbed by the behavior of his colleagues and superiors, Singh decides to leave QH. He
interviews with QH’s largest competitor, Algos-R-Us (ARU). During his interview, Singh
shows them a proprietary model – StockStar – that he has been developing for the past two
years in his spare time (nights and weekends). In addition to back-testing the model, Singh
has used StockStar to manage his personal portfolio and the portfolios of his family in order
to generate actual performance results. Impressed by Singh and his model, ARU not only
offers him a job but also offers to pay him a special licensing fee for the use of his StockStar
model. Singh accepts and resigns from QH.
At ARU Singh provides the following information about StockStar. This information will be
incorporated into a new marketing brochure that will be mailed to current and prospective
clients.
StockStar Model Performance: Actual and Back-Tested Returns*
Model’s Benchmark Return Excess Return (%)
Performance (%) (%)
Back-tested returns
2012 10.5 6.5 4.0
2013 –2.5 –6.0 3.5
2014 20.3 25.0 –4.7
2015 0.5 –6.8 7.3
Actual returns
2016 8.5 2.2 6.3
2017 12.0 –0.5 12.5
* The benchmark return is based on the S&P/ASX 200 index. Note that past performance
does not guarantee future results
In the brochure, Singh states: “This model has been used to manage real portfolios over the

past two years and has outperformed its benchmark in both years. In back-tests, the model
has outperformed its benchmark in three out of four years.”
Case questions
1. Does Singh violate the CFA Institute Code and Standards by using the ATM model
exclusively for QH’s institutional clients?
No, III (B): Duties to Clients allows us to use different investment models for different types
of clients. Each client has unique needs, investment criteria, and investment objectives, so
not all investment opportunities are suitable for all clients.
2. Prior to the performance concerns voiced by QH’s institutional clients, did Singh
violate the CFA Institute Code and Standards in the updating of the ATM model
components?
Yes, Singh violated Standard V(A): Investment Analysis, Recommendations, and Actions–
Diligence and Reasonable Basis. Although the Alpha and Risk components are updated
quarterly, he reviews the Optimizer, which links the two prior components, and the overall
model itself on only an annual basis. All components should have been tested with the
same quarterly frequency.
3. According to the CFA Institute Code and Standards, what is the next action
(Dissociate from the firm; Contact the firm’s clients; Contact senior management)
that Singh should take following his conversation with Ringfield about the model
error?
Singh should contact senior management before dissociating himself from QH or contacting
QH clients. According to Standard I(A) Professionalism–Knowledge of the Law, “The first
step should be to attempt to stop the behavior by bringing it to the attention of the
employer through a supervisor or the firm’s compliance department. If this attempt is
unsuccessful, then members and candidates have a responsibility to step away and
dissociate from the activity.” Contacting the firm’s clients directly is not a permitted
intermediate step under Standard I(A).
4. Did Singh violate the CFA Institute Code and Standards by not immediately fixing
the error in the ATM model?
Yes, Singh violated Standard III(A): Duties to Clients–Loyalty, Prudence, and Care by not
fixing the error in the ATM model immediately.
5. Did Ringfield violate the CFA Institute Code and Standards when talking with
clients about their portfolios’ underperformance?
Yes, Ringfield violated Standard I(C): Professionalism–Misrepresentation. By attributing
the underperformance of client portfolios to market volatility, she is not telling them the
real reason for the underperformance.

6. Did Singh violate the CFA Institute Code and Standards with respect to his duties to
his employer, QuantHouse, in developing his StockStar model?
No, Singh did not violate Standard IV(A): Duties to Employers–Loyalty since he developed
this model in his spare time and used it to manage only his personal and family portfolios.
7. In the table that Singh provides to the marketing department, does he violate the
CFA Institute Code and Standards?
No, Standard III(D): Duties to Clients–Performance Presentation has not been violated.
Singh has presented both the actual and back-tested performance of the model and clearly
distinguished between the two. He has also noted that past performance does not
guarantee future results.
JR and Associates
Case facts
Jacobs, Riccio, and Associates (JRA) is a global investment advisory firm. Benjamin Jacobs,
CFA, and Andrew Riccio, CFA, founded JRA 10 years ago. Prior to establishing the firm,
Jacobs worked as a lawyer for Brightman Partners (BP) and Riccio worked as a Certified
Public Accountant for Earnest & Olds (E&O). Kathy Parker, CFA, joined the firm as the third
senior partner two years after it was founded. Previously, she had worked for the Frontline
Group (FG), a broker/dealer. JRA acquires most of its clients through referral arrangements
put in place by the three senior partners.
Jacobs has a fee-sharing arrangement with his former colleagues at Brightman Partners
(BP) when they refer clients to JRA. The annual investment fee stated in JRA’s marketing
brochure is higher than the fee most of its clients pay because Jacobs offers a discount on
the investment fee to clients who are referred by BP lawyers. JRA shares a portion of the
client’s’ annual investment advisory fee with the referring lawyer. The lawyers at BP
disclose this fee-sharing arrangement with the clients that they refer to JRA. JRA discloses
all of this information in the JRA investment management agreement that individuals sign
at the time they become clients.
Riccio offers a similar fee discount and sharing arrangement to accountants at his previous
firm, Earnest & Olds (E&O), who refer their clients to JRA. To assist JRA advisers in
developing more realistic and accurate investment policy statements, the accountants at
E&O provide a copy of their referred client’s tax returns to the client’s JRA adviser after
they open an account at JRA. This step allows JRA advisers to “know their client” better and
provides greater transparency into their client’s financial condition. In return, JRA advisers
provide their clients’ quarterly account statements to their E&O accountants to help with
their tax planning and year-end tax preparation.
Kathy Parker has a somewhat different referral arrangement in place with the Frontline
Group. Frontline’s brokerage unit refers all of its small institutional clients that are looking

for investment management to JRA. In return, all of the trading from these accounts
continues to be executed through Frontline’s broker/dealer. Because Frontline continues to
provide “best price and best execution” to these clients, Parker believes no additional client
disclosures are necessary because client trading is unaffected.
Since starting JRA, Jacobs and Riccio have developed a close relationship with Tim Carroll,
an independent consultant they met at a networking event. Over the years, Carroll has been
instrumental in JRA’s success by referring several of his pension fund clients to the firm
because of the firm’s outstanding performance record and superior client service. To thank
Carroll for all of his hard work on JRA’s behalf (regardless of whether Carroll’s pension
fund clients actually hire JRA), Jacobs and Riccio each make sizable annual donations to
Carroll’s Children’s Charity, a non-profit organization Carroll created to benefit orphans.
These donations are not disclosed to the pension funds referred by Carroll who become
JRA clients.
Recently, JRA hired Mufid Othan, an investment adviser and CFA charterholder who
previously worked at JRA’s largest competitor, Sack International. To attract Othan and his
large “book of clients,” JRA offered him $500 for each client he “brought over” from Sack.
While at Sack, Othan was allowed to connect with all of his clients through his personal
social media platforms. When Othan tendered his resignation from Sack, he was
immediately escorted out of the building. Othan spent the following weekend contacting all
of his clients via social media to tell them about his resignation and to encourage them to
join him at JRA. He did not disclose to them, however, that he was being paid $500 for each
client he brought over from Sack.
A few weeks after beginning work at JRA, Othan hired Zane Ode, a recent college graduate,
who recently found out she had passed Level III of the CFA Program examination. After
hearing the good news about her success with Level III, Ode posted the following
comments in a CFA candidate chatroom:
Comment 1 “I can’t believe I passed the exam; the ethics questions were super hard.”
Comment 2 “Wow, I scored above the Minimum Performance Score (MPS) on
derivatives. I still don’t know what answer was right for the two-part contango–
backwardation question.”
Comment 3 “The graders must have been quite lenient in grading my answers to the
constructed response questions.”
Ode now has three and a half years of experience in the investment industry. Nevertheless,
Othan has already made a habit of introducing her to current and prospective clients as the
firm’s “newest CFA,” and Ode has said nothing to correct him.
Case questions
1. Does Jacobs violate the CFA Institute Code and Standards by offering his referral

clients a lower investment advisory fee than the one quoted in JRA’s marketing
brochure?
No. According to Standard III (B): Duties to Clients–Fair Dealing, members and candidates
may provide more personal, specialized, or in-depth service to clients who are willing to
pay for premium services through higher management fees or higher levels of brokerage.
2. Does Jacobs violate the CFA Institute Code and Standards in his disclosure of
referral arrangements to his clients?
Yes, Jacobs is in violation of Standard VI (C): Conflicts of Interest–Referral Fees. The
disclosure does not occur until the time the individual signs the investment management
agreement, which is too late.
3. Do JRA advisers violate the CFA Institute Code and Standards by sharing client
information with the accountants at E&O?
Yes, Standard III(E): Duties to Clients–Preservation of Confidentiality is violated by not
obtaining client approval in advance of sharing their information between each firm.
4. Has Parker violated the CFA Institute Code and Standards in her referral
arrangement with Frontline Group?
Yes, Parker has violated Standard VI (C): Conflicts of Interest–Referral Fees. Regardless of
whether Frontline provides “best price” and “best execution” or whether the execution of
client trades remains unchanged by Frontline, Parker must still disclose the referral
arrangement to her clients.
5. Did Jacobs and Riccio violate the CFA Institute Code and Standards by making
annual donations to Carroll’s Children’s Charity?
Yes, Jacobs and Riccio have violated Standard I(B): Professionalism–Independence and
Objectivity. The donations made by Jacobs and Riccio give Carroll an incentive to refer
potential clients to JRA and at the very least give the perception that Carroll’s objectivity
and independence have been compromised.
6. Did Othan violate the CFA Institute Code and Standards by contacting his Sack
International clients via social media after leaving Sack?
No. In this case, Sack allowed Othan to use his personal social media platforms to connect
with clients. In addition, he did not contact his former clients via social media to inform
them about his departure until after he resigned from Sack. Therefore, he did not violate
Standard IV(A): Duties to Employers–Loyalty.
7. Did Othan violate the CFA Institute Code and Standards by not disclosing to clients
that he was receiving $500 for each client that he brought over to JRA from Sack?
No. The $500 does not have to be disclosed to clients because it is not a referral fee or
additional compensation, and it does not create a conflict of interest with his employer,

clients, or prospective clients.

8. Which of the comments Ode posted in the CFA candidate chatroom violated the
CFA Institute Code and Standards?
Ode’s comment 2 violated Standard VII(A): Responsibilities as a CFA Institute Member or
CFA Candidate–Conduct as Participants in the CFA Institute Programs. In this case, Ode
disclosed specific details of questions appearing on the exam.
9. Did Othan violate the CFA Institute Code and Standards in his description of Ode?
Yes. Ode is not yet a CFA charterholder, and in referencing her as the firm’s “newest CFA,”
Othan is misrepresenting Ode.
Magadi Asset Management

Magadi Asset Management (Magadi) is a global investment management firm based in
Nairobi, Kenya. Frederick Omondi, CFA, is Magadi’s president and chief investment officer.
The traders at the main trading desk handle the orders of Magadi’s institutional or retail
clients. Last year, Omondi also established a proprietary trading desk at Magadi. Both
trading desks are located on the same floor at Magadi’s headquarters. Omondi has told
clients that information concerning their orders and business affairs is kept confidential.
Omondi recently secured a large mandate from a sovereign wealth fund. To win this
mandate, Omondi hired, as a ‘sub-advisor’ a business development agent. Despite having
very limited experience as a financial consultant, the agent had a number of close
relationships with senior managers at the sovereign wealth fund. As a thank you for being
awarded the mandate, Omondi made donations to the favorite charities of the sovereign
wealth fund’s top management, as he had promised during the due diligence process.
Three years ago, Magadi launched the Pan Africa Frontier Fund (PAFF), a non-listed equity
unit trust with the following investment mandate:
 80% of the companies in the portfolio to be traded on at least one of the 17
securities exchanges operating within Africa;
 the portfolio be invested in a minimum of eight countries at all times;
 no more than 30% of the portfolio’s value can be invested in any single country;
 no more than 10% of the portfolio’s value can be invested in cash and cash
equivalents; and
 no single security can account for more than 15% of the portfolio’s value.
Since its launch, PAFF has significantly underperformed its peers.
Omondi recently hired Bukenya Kirabo, CFA, to take over management of PAFF and to
improve its performance. Kirabo makes the following changes:
Change 1: Increases the maximum level of cash and cash equivalents to 15% of the

portfolio.
Change 2: Increases the portfolio’s geographic diversification from 11 countries to
13.
Change 3: To increase accountability for PAFF’s performance, Kirabo himself will
now be making all buy and sell decisions for PAFF.
Kirabo next meets with the marketing department to discuss PAFF’s new sales campaign.
He asks the marketing team to include all of the mandate changes in the new brochure that
will be distributed to prospective clients. But he tells them that because the mandate
changes are relatively trivial, there is no need to inform existing clients. He also asks them
to include the five-year investment performance he achieved at his previous employer.
However, he asks them not to state where the performance was earned because his
previous employer is a direct competitor of Magadi.
PAFF currently owns 9% of the common stock of Mtume’s, a mining company. Kirabo has
been reducing the fund’s holdings in Mtume because of the company’s declining revenues
and profits. This morning, Kirabo speaks with Olivia Moroka, Mtume’s chief financial
officer. During their conversation, Moroka tells Kirabo, “You may want to stop selling your
shares of Mtume, because our board of directors just received a very attractive all-cash
offer of BWP500 million (Botswana pula) to purchase one of our mining subsidiaries.
Although nothing is definite, the board will be meeting next week to vote on the offer.”
After getting off the phone with Moroka, Kirabo calls an analyst and asks her to incorporate
the expected subsidiary sales price into her financial model of Mtume. The output from her
revised model indicates that the sale proceeds will significantly enhance Mtume’s credit
standing and its ability to reinstitute shareholder cash distributions on an earlier-than-
expected schedule and in larger-than-expected amounts. When the analyst tells Kirabo
about her findings, Kirabo immediately calls the proprietary and main traders to tell them
to start buying “any and all” shares of Mtume. He then calls Omondi and tells him about his
conversation with Moroka. After Omondi gets off the phone with Kirabo, Ormondi calls his
broker and purchases shares in Mtume for his personal account and the family accounts
that he controls.
Case questions
1. By allowing customer order information to be known to the traders on the
proprietary desk, did traders on the main trading desk most likely violate the CFA
Institute Code and Standards?
Yes, Standard III(E): Duties to Clients–Preservation of Confidentiality has been violated.
The sharing of office space such that the proprietary traders can see the screens of the
main traders is inappropriate because it allows confidential client information to be
disclosed to proprietary traders who did not need to know the information.

2. Did Omondi most likely violate the CFA Institute Code and Standards in
supervising the employees in the two trading desks?
Yes, Omondi violated Standard IV(C): Duties to Employers–Responsibilities of Supervisors.
Omondi failed to establish effective policies and procedures reasonably designed to
prevent traders on the proprietary dealing desk from obtaining confidential customer
information.
3. Did Omondi violate the CFA Institute Code and Standards when dealing with the
sovereign wealth fund’s top managers?
Yes, Omondi violated Standard I(B): Professionalism–Independence and Objectivity. It is
clear from the facts of this case that Omondi is not hiring a true sub-adviser but instead
paying locally connected officials to secure access for the sovereign wealth fund’s
investment. The donations made to charity also impair the top management’s
independence and objectivity.
4. According to the CFA Institute Code and Standards, which of the changes in the
PAFF Fund does Kirabo not have to disclose?
Change 2 is not required to be disclosed because, by increasing the country exposure to 13
nations, Kirabo is still within the 80% stated mandate. The investment process has not
fundamentally changed.
5. Does Kirabo most likely violate the CFA Institute Code and Standards by including
his prior performance in the PAFF marketing brochure?
Yes, because the brochure should have stated the name of the firm where he earned prior
performance.
6. According to the CFA Institute Code and Standards, who must, Kirabo most likely
inform of the material changes related to the PAFF?
According to Standard V(B): Investment Analysis, Recommendations, and Actions–
Communication with Clients and Prospective Clients, Kirabo must disclose to both the
current and prospective clients the fund mandate change and the change in the investment
process.
7. Did Kirabo most likely violate the CFA Institute Code and Standards by purchasing
additional shares of Mtume?
Yes, Kirabo violated Standard II(A): Integrity of Capital Markets–Material Nonpublic
Information. Although the information is not definite or officially accepted by the board, its
source, substance, and specificity are enough to make the information material.
8. Did Omondi most likely violate the CFA Institute Code and Standards by
purchasing shares for his personal and family accounts?
Yes. By purchasing shares informed by material nonpublic information, Omondi violated

Standard II(A).
Edvard Stark
Case facts
Edvard Stark, CFA is a private client advisor for Eyearne Bank. He recently became
interested in cryptocurrencies. Cryptocurrencies are held in online wallets. Account-
holders can earn additional cryptocurrencies tokens through an activity called ‘mining’.
After studying top cryptocurrencies over two weekends, Stark decides to go with a newer
digital currency called Meerine. To limit his risk of being wrong, Stark decides to give a buy
recommendation to only a few small clients. He recommends a 1% position in Meerine to
these clients.
Stark continues learning more about cryptocurrencies. He attends conferences and
workshops. He starts mining Meerine’s currency by running the mining software as a
background process on his home computer for several months. This provides him with a
way to supplement his salary. He believes that he is now competent in his understanding of
cryptocurrencies and their underlying technology. During this time, Meerine’s price
continues to rise strongly.
Stark decides to recommend a 3% Meerine position for all clients. His clients know very
little about cryptocurrencies. They have a few questions and no objections. Stark, therefore,
feels that his recommendation has been well received. He offers his larger clients the
opportunity to buy Meerine tokens directly from him.
Case discussion
Standard III (B) - Duties to Clients: Fair Dealing
By offering only his large clients the opportunity to buy Meerine tokens from him, Stark has
violated this standard.
Action required:
Stark should either (1) make this offer to all his clients and then allocate tokens in
proportion to the clients’ investments or (2) not make this offer to any client.
The firm can adopt a policy where such transactions are executed as a block trade, and
allocations are made on a pro-rata basis, where everyone receives the same execution
price.
Standard III (C) – Duties to Clients: Suitability
Starks initial recommendation of 1% to only his smallest clients and later recommendation
of 3% to all clients is a violation of this standard.
Action required:
Stark should consider each client’s risk tolerance, goals and objectives to determine if that

client can invest in Meerine and the appropriate level of exposure for that client.
Standard IV (B) – Duties to Employers: Additional Compensation Arrangements
By mining Meerine, Stark is supporting a service that is competitive with his employer
Eyearne bank. This creates a conflict of interest. Also, he is earning additional
compensation through mining and has not disclosed this fact to his employer. Therefore,
standard IV (B) is violated.
Action required:
Stark needs to disclose to his employer, his intention to mine Meerine and the potential
earnings expected from this activity. He needs to obtain written consent before beginning
this activity.
Standard V (A): Investment Analysis, Recommendations, and Actions: Diligence and
Reasonable Basis
Although Stark had done some research, he was still in the learning process when he made
the 1% buy recommendation. Therefore, standard V (A) was violated.
Action required:
Stark should develop a written report detailing the background information and decision
framework that support his recommendation.
The firm can create an ‘approved list’ for securities. Only securities on this list can be
recommended or purchased. Securities can be added to this list only after they have been
approved by an investment committee.
Standard VI (A) Conflict of Interest: Disclosure of Conflicts
Stark’s mining of Meerine and his recommendation that clients invest in Meerine is a
conflict of interest. If Meerine appreciates due to his client’s purchases; this will directly
benefit Stark. Stark’s lack of full disclosure is a violation of Standard VI (A).
Action required:
Stark should clearly disclose to his clients and employer the conflict of interests. He should
also determine suitable alternative cryptocurrencies for clients who are uncomfortable
with this conflict.
Subath Agarway
Case facts
Subath Agarway has recently joined CrowdWisdom as vice president, and he is in charge of
due diligence. CrowdWisdom offers an online matching platform that matches investors
with startup companies in need of capital. Agarway’s role is to identify suitable companies
to list on the online matching platform.
CrowdWisdom’s cofounders Craig and Stephane have created a subgroup of investors

called ‘Investor Club’ by selecting the most active investors on the online platform. This
subgroup has access to market intelligence research in addition to the research provided
on the CrowdWisdom platform.
Agarway’s due diligence process consists of a two-step process he developed at his
previous company FunderWise. He is confident in his process and has personally invested
in several FunderWise listed companies using this approach.
Agarway has identified a very promising start-up company called Deko. Most of Deko’s
customers are pre-teens and teenagers. CrowdWisdom could approach these customers for
future investor funding activities. The company’s strategy is to market its crowdfunded
shares through email communications. The email states that the offer is available to adults
over the age of 18.
Agarway’s responsibilities have increased over time. His stack of applications for review
grows to 300 companies. To meet CrowdWisdom’s aggressive growth goals, the company
co-founders ask Agarway to target a 10% acceptance rate. They suggest that he should find
ways to reduce the time spent on each application. They also recommend the acceptance of
two companies whose founder’s they met at a recent conference.
Case discussion
Standard I (A) - Professionalism: Knowledge of the Law
Agarway should review the global rules governing online marketing to Deko’s teen and
preteen customers. They may be at risk of prosecution because it is illegal in many
countries to collect information on such individuals over the internet without first
obtaining parental permission.
Action required:
CrowdWisdom should have legal counsel available to review planned additions to the
platform to ensure that the company’s strategy is not in conflict with relevant laws.
Standard I(B) - Professionalism: Independence and Objectivity
Pressure from the founders to increase the number of listings and reducing the review
timeframe is likely to compromise Agarway’s independence and objectivity.
Action required:
They need to come up with a process that is acceptable to both Agarway and the founders.
Standard VI(A) - Conflicts of Interest: Disclosure of Conflicts
Two possible conflicts of interest in this case are:
 Investor Club getting access to additional market intelligence research.
 Agarway’s personal investments in several companies that could be listed on
CrowdWisdom’s platform in the future or may compete with the platform listed
companies.

Action required:
The benefits of the Investor Club and the requirements to join this club should be disclosed
to all investors on the CrowdWisdom platform.
Aggarway’s personal investments should be disclosed to his supervisor and
CrowdWisdom’s compliance officer. If the firms are listed on CrowdWisdom’s platform,
then disclosures should be made to the platform users as well.

High Yield Notes

Uploaded by

Copyright:

Available Formats

High Yield Notes

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

High Yield Notes

Uploaded by

Copyright:

Available Formats

Quantitative Methods 2023 Level II High Yield Notes

LM01 Basics of Multiple Regression and Underlying Assumptions

The basics of multiple linear regression

Assumptions underlying multiple linear regression

© IFT. All rights reserved 1

implies the regression residuals are uncorrelated across observations.

© IFT. All rights reserved 2

LM02 Evaluating Regression Model Fit and Interpreting Model Results

Testing joint hypotheis for coefficients

© IFT. All rights reserved 1

Forecasting using multiple regression

© IFT. All rights reserved 2

LM03 Model Misspecification

© IFT. All rights reserved 1

© IFT. All rights reserved 2

computing ‘robust standard errors’.

© IFT. All rights reserved 3

LM04 Extensions of Multiple Regression

where: k is the number of independent variables

© IFT. All rights reserved 1

Dummy variables in multiple regression

Exhibit 11, Panel B, shows the effect of a slope dummy variable.

© IFT. All rights reserved 2

Logistic regression model

© IFT. All rights reserved 3

© IFT. All rights reserved 4

LM05 Time-Series Analysis

Linear vs log-linear trend models

Requirements for a time series to be covariance stationary

Autoregressive (AR) models

© IFT. All rights reserved 1

Autocorrelations of the residuals

© IFT. All rights reserved 2

© IFT. All rights reserved 3

Autoregressive conditional heteroskedasticity (ARCH)

Working with two time series

Selecting an appropriate time-series model

© IFT. All rights reserved 4

© IFT. All rights reserved 5

© IFT. All rights reserved 6

LM06 Machine Learning

© IFT. All rights reserved 1

Supervised machine learning algorithms

Unsupervised machine learning algorithms

© IFT. All rights reserved 2

Neural networks, deep learning nets, and reinforcement learning

© IFT. All rights reserved 3

LM07 Big Data Projects

Data Prep & Wrangling

© IFT. All rights reserved 1

 Duplication errors: Duplicate observations are present.

© IFT. All rights reserved 2

© IFT. All rights reserved 3

 Converting numbers into tokens: To preserve the meaning conveyed by numbers of

© IFT. All rights reserved 4

Evaluating the Fit of a ML algorithm

© IFT. All rights reserved 5

LM01 Intercorporate Investments

Investments in financial assets: IFRS 9

© IFT. All rights reserved 1