2019 Fin Econ

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

HARARE INSTITUTE OF TECHNOLOGY

SCHOOL OF BUSINESS AND MANAGEMENT SCIENCES


DEPARTMENT OF FINANCIAL ENGINEERING

B. TECH DEGREE IN FINANCIAL ENGINEERING

BFE 2111: FINANCIAL ECONOMETRICS I

TIME: 3 HOURS DATE: NOV/DEC 2019

INSTRUCTIONS TO CANDIDATES:

 The paper has six (6) questions


 Choose any four (4) questions
 Each question carries 25 marks
 The artificial and natural persons in this paper are intended to be fictitious

ADDITIONAL MATERIALS

 Statistical tables

Page 1 of 6
Question One

a) A financial econometrician estimates the following model for stock market


returns, which is as follows:
Y t = 0.638+ 0.402X 2 t −0.891X 3t , R 2 = 0.96, Adjusted R 2 = 0.89

(0 .436) (0.291) (0.763)


He thinks that there may be a problem with the model.
i) By calculating the t-ratios, and considering their significance and
simultaneously examining the value of R 2 or otherwise, suggest what
the problem might be. [5]
ii) How might you go about solving the perceived problem? [4]
b) A fellow financial engineering student suspects that the residuals of her model
might be autocorrelated. Explain the steps involved in testing this theory using
the Durbin–Watson (DW) test. [5]

c) In each of the models below, show how it is possible to apply ordinary least
squares estimation method.


i. Yt  AX t e ut [3]


ii. Yt     ut [3]
Xt

d) Distinguish between Type I and Type II errors in hypothesis testing. [5]

Question Two

a) Consider a bivariate relationship between an organisation’s advertising


expenditure and its sales revenue.

i. Show a sketch of the scatter diagram that likely represents this


relationship. [1]

ii. Superimpose a straight line that best generalises the bivariate


relationship. [1]

iii. Write the regression equation that fits the generalisation in (ii) above.

Page 2 of 6
[2]

iv. Derive estimators of the two unknown parameters of the regression


equation in (iii) above. [9]

b) You have estimated the following ARMA(1,1) model for some time series data
y t = 0.036+0.69y t 1 +0.42u t 1 +u t . Suppose that you have data for time t−1,

i.e. you know that y t 1 = 3.4, and ˆ u t 1 =−1.3

i) Obtain forecasts for the series y for times t, t+1, and t+2 using the
estimated ARMA model. [6]

ii) If the actual values for the series turned out to be −0.032, 0.961, 0.203
for t, t+1, t+2, calculate the out-of-sample mean squared error.
[6]

Question Three

a) Show why the Durbin Watson test statistic is 0≤ DW ≤4. [4]


b) If estimators determined by Ordinary Least Squares method have a number of
desirable properties, then they are known as BLUE. Discuss. [6]
c) Outline how you would conduct the Lagrange Multiplier (LM) test for
heteroscedasticity. [4]
d) Consider the following autocorrelation and partial autocorrelation coefficients
estimated using 500 observations for a weakly stationary series, y t :

Lag acf pacf


1 0.307 0.307
2 −0.013 0.264
3 0.086 0.147
4 0.031 0.086
5 −0.197 0.049

i. Determine which, if any, of the acf and pacf coefficients are significant
at the 5% level. [5]

Page 3 of 6
ii. Use both the Box–Pierce and Ljung–Box statistics to test the joint null
hypothesis that the first five autocorrelation coefficients are jointly zero.
[6]
Question Four

a) Explain, with the use of equations or mathematical notation, the following


processes:
i. Autoregressive. [3]
ii. Moving Average. [3]
iii. Autoregressive Moving Average. [3]
b) A monthly model of asset returns had been estimated for 2000 to 2015 and
the residuals plotted. The Jarque-Bera test has also been conducted on the
residuals and the normality assumption has been rejected. From the plot of
the residuals it has been observed that a large outlier is present for December
2010. Illustrate how error normality can be improved in this circumstance. [7]

c) Consider an estimated function Y = 20.3 + 0.5091X

(14.38) (0.2561)

Given that T=22 and using the 5% level of significance, test the hypothesis
that the slope of this function is equal to 1 against a two sided alternative
using both the test of significance and the confidence interval approach. [9]

Question Five

a) With the aid of technical notations, explain the following time series concepts:

i. Strictly stationary process [3]


ii. Weakly stationary process [3]
iii. White noise process [3]
b) Justify use of univariate time series models as opposed to structural models in
financial and economic modelling. [4]
Page 4 of 6
c) Explain the consequence of dropping the intercept term in any regression
model. [2]

d) Describe the few circumstances under which regression could be forced to be


done through the origin. [3]

e) Suppose that it is now January 1993. Consider the following regression for the
standard CAPM β for the returns on a stock
rgt    rMt   t

where rgt and rMt are excess returns on Glaxo shares and on a market

portfolio, respectively. Suppose that you are interested in estimating beta


using monthly data from 1981 to 1992, to aid a stock selection decision.
Another researcher expresses concern that the October 1987 stock market
crash fundamentally altered the risk–return relationship. Given that the model
for each sub-period is:
1981M1-1987M10:

rgt  0.24  1.2rMt , T=82 and RSS 1 =0.03555

1987M11-1992M12:

rgt  0.68  1.53rMt , T=62 and RSS 2 =0.00336

1981M1-1992M12:

r gt  0.39  1.37rMt , T=144 and RSS=0.0434

Test the above conjecture using a Chow test of parameter stability. [7]

Question Six

a) Use the standardised third and fourth moments of a distribution to


diagrammatically illustrate the distinction between a mesokurtic, leptokurtic
and platykurtic distributions. [6]
b) Illustrate the distinction between recursive and rolling window forecasts. [4]

c) Distinguish between the population and the sample regression function. [3]

Page 5 of 6
d) Given that the price of an asset follows a Moving Average (MA) process of the
following form: Pr icet    1 t 1   2 t 2   3 t 3   t

Show that f t , 4   [6]

e) The following is an output for tests on residuals from testing whether inflation
was indeed a monetary phenomenon in Zimbabwe between 1975 and 2005.

Comment on whether the residuals from this test were normally distributed or
not. [6]

THE END

Page 6 of 6

You might also like