7basic Econometrics Chapter Vi

Basic Econometrics
Chapter 6
EXTENSIONS OF THE
TWO-VARIABLE LINEAR
REGRESSION MODEL
1
Prof. Himayatullah May 2004
Chapter 6
EXTENSIONS OF THE TWO-VARIABLE
LINEAR REGRESSION MODELS
6-1. Regression through the origin

 The SRF form of regression:
 Yi = b^2X i + u^ i (6.1.5)
 Comparison two types of regressions:

* Regression through-origin model and
* Regression with intercept
2
Chapter 6

Comparison two types of regressions:
b^2 = SXiYi/SX2i (6.1.6) O

b^2 = Sxiyi/Sx2i (3.1.6) I
var(b^2) = s2/ SX2i (6.1.7) O
var(b^2) = s2/ Sx2i (3.3.1) I
s^2 = S(u^i)2/(n-1) (6.1.8) O
s^2 = S(u^i)2/(n-2) (3.3.5) I
3
Chapter 6
EXTENSIONS OF THE TWO-VARIABLE LINEAR
REGRESSION MODELS
 r2 for regression through-origin model
Raw r2 = (SXiYi)2 /SX2i SY2i (6.1.9)
 Note: Without very strong a priory expectation, well
advise is sticking to the conventional, intercept-
present model. If intercept equals to zero
statistically, for practical purposes we have a
regression through the origin. If in fact there is an
intercept in the model but we insist on fitting a
regression through the origin, we would be
committing a specification error
4
Chapter 6

 Illustrative Examples:
1) Capital Asset Pricing Model - CAPM (page 156)
2) Market Model (page 157)
3) The Characteristic Line of Portfolio Theory
(page 159)
5
Chapter 6
EXTENSIONS OF THE TWO-VARIABLE LINEAR
REGRESSION MODELS
6-2. Scaling and units of measurement
 Let Yi = b^1 + b^2Xi + u^ i (6.2.1)

 Define Y*i=w 1 Y i and X*i=w 2 X i then:
 b*^2 = (w1/w2) b^2 (6.2.15)
 b*^1 = w1b^1 (6.2.16)
 s*^2 = w12s^2 (6.2.17)
 Var(b*^1) = w21 Var(b^1) (6.2.18)
 Var(b*^2) = (w1/w2)2 Var(b^2) (6.2.19)
 r2xy = r2x*y* (6.2.20)
6
Chapter 6
6-2. Scaling and units of measurement
 From one scale of measurement, one can derive the results
based on another scale of measurement. If w1= w2 the
intercept and standard error are both multiplied by w1. If
w2=1 and scale of Y changed by w1, then all coefficients and
standard errors are all multiplied by w1. If w1=1 and scale of
X changed by w2, then only slope coefficient and its standard
error are multiplied by 1/w2. Transformation from (Y,X) to
(Y*,X*) scale does not affect the properties of OLS
Estimators
 A numerical example: (pages 161, 163-165)
7
6-3. Functional form of regression model
 The log-linear model

 Semi-log model
 Reciprocal model
8
6-4. How to measure elasticity
The log-linear model
 Exponential regression model:
 Yi= b1Xi b2 e u i (6.4.1)
By taking log to the base e of both side:
 lnYi = lnb1 +b2lnXi + ui , by setting lnb1 = a =>
 lnYi = a +b2lnXi + ui (6.4.3)
(log-log, or double-log, or log-linear model)
This can be estimated by OLS by letting
 Y*i = a +b2X*i + ui , where Y*i=lnYi, X*i=lnXi ;
b2 measures the ELASTICITY of Y respect to X, that is,
percentage change in Y for a given (small) percentage
change in X.
9
6-4. How to measure elasticity
The log-linear model

The elasticity E of a variable Y with
respect to variable X is defined as:
E=dY/dX=(% change in Y)/(% change in X)
~ [(Y/Y) x 100] / [(X/X) x100]=
= (Y/X)x (X/Y) = slope x (X/Y)
 An illustrative example: The coffee

demand function (pages 167-168)
10
6-5. Semi-log model:
Log-lin and Lin-log Models
 How to measure the growth rate: The log-lin model

 Y t = Y0 (1+r) t (6.5.1)
 lnYt = lnY0 + t ln(1+r) (6.5.2)
 lnYt = b1 + b2t , called constant growth model (6.5.5)
where b1 = lnY0 ; b2 = ln(1+r)
 lnYt = b1 + b2t + ui (6.5.6)
 It is Semi-log model, or log-lin model. The slope
coefficient measures the constant proportional or
relative change in Y for a given absolute change in the
value of the regressor (t)
 b2 = (Relative change in regressand)/(Absolute change
in regressor) (6.5.7)
11
 Instantaneous Vs. compound rate of growth
 b2 is instantaneous rate of growth
 antilog(b2) – 1 is compound rate of growth
The linear trend model
 Yt = b1 + b2t + ut (6.5.9)
 If b2 > 0, there is an upward trend in Y
 If b2 < 0, there is an downward trend in Y
 Note: (i) Cannot compare the r2 values of
models (6.5.5) and (6.5.9) because the
regressands in the two models are different,
(ii) Such models may be appropriate only if a
time series is stationary.
12
 The lin-log model:
 Yi = b1 +b2lnXi + ui (6.5.11)
 b2 = (Change in Y) / Change in lnX =
(Change in Y)/(Relative change in X) ~
(Y)/(X/X) (6.5.12)
 or Y = b2 (X/X) (6.5.13)
 That is, the absolute change in Y equal
to b2 times the relative change in X.
13
6-6. Reciprocal Models:
The reciprocal model:

 Yi = b1 + b2( 1/Xi ) + ui (6.5.14)
 As X increases definitely, the term
b2( 1/Xi ) approaches to zero and Yi
approaches the limiting or asymptotic value
b1 (See figure 6.5 in page 174)
 An Illustrative example: The Phillips Curve
for the United Kingdom 1950-1966
14
6-7. Summary of Functional Forms
Table 6.5 (page 178)
Model Equation Slope = Elasticity =

dY/dX (dY/dX).(X/Y)
Linear Y = b1 + b2 X b2 b2(X/Y) */
Log-linear lnY = b1 + b2 lnX b2 (Y/X) b2

(log-log)
Log-lin lnY = b1 + b2 X b2 (Y) b2 X */
Lin-log Y = b1 + b2 lnX b2(1/X) b2 (1/Y) */
Reciprocal Y = b1 + b2 (1/X) - b2(1/X2) - b2 (1/XY) */
15
6-7. Summary of Functional Forms
 Note: */ indicates that the elasticity
coefficient is variable, depending on the
value taken by X or Y or both. when no X
and Y values are specified, in practice, very
often these elasticities are measured at the
mean values E(X) and E(Y).
-----------------------------------------------
6-8. A note on the stochastic error term
6-9. Summary and conclusions
(pages 179-180)
16

7basic Econometrics Chapter Vi

Uploaded by

Copyright:

Available Formats

7basic Econometrics Chapter Vi

Uploaded by

Document Information

Original Description:

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

7basic Econometrics Chapter Vi

Uploaded by

Copyright:

Available Formats

Basic Econometrics

6-1. Regression through the origin

 Comparison two types of regressions:

6-1. Regression through the origin

b^2 = SXiYi/SX2i (6.1.6) O

6-1. Regression through the origin

 Let Yi = b^1 + b^2Xi + u^ i (6.2.1)

 The log-linear model

The log-linear model

 An illustrative example: The coffee

 How to measure the growth rate: The log-lin model

The reciprocal model:

Model Equation Slope = Elasticity =

Log-linear lnY = b1 + b2 lnX b2 (Y/X) b2

You might also like