Scribd
Upload
268 views

Exercise 1. Open GRETL and Capm - GDT Data File.: Coefficient Std. Error T-Ratio P-Value

This document summarizes steps taken to analyze stock return data in GRETL: 1. The excess return of GM stock was calculated by subtracting the risk-free rate from GM's stock return, which was defined as the log difference of monthly stock prices. 2. An OLS regression model was estimated with excess GM return as the dependent variable and market excess return as the independent variable. The model found a positive relationship between excess returns, though the R-squared was low. 3. Residual analysis showed the residuals had a mean of approximately 0, indicating the regression assumptions were met. Interpretation of the coefficients and test statistics was also provided.

Uploaded by

Brian Tran
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
268 views

Exercise 1. Open GRETL and Capm - GDT Data File.: Coefficient Std. Error T-Ratio P-Value

This document summarizes steps taken to analyze stock return data in GRETL: 1. The excess return of GM stock was calculated by subtracting the risk-free rate from GM's stock return, which was defined as the log difference of monthly stock prices. 2. An OLS regression model was estimated with excess GM return as the dependent variable and market excess return as the independent variable. The model found a positive relationship between excess returns, though the R-squared was low. 3. Residual analysis showed the residuals had a mean of approximately 0, indicating the regression assumptions were met. Interpretation of the coefficients and test statistics was also provided.

Uploaded by

Brian Tran
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Exercise 1. Open GRETL and capm.gdt data file.

1. Compute the excess return of GM stock


Excess return of GM stock is the excess between the return of GM stock and return of
risk free assets
(1) Exr_GM=r_GM-rf
(2) R_GM=100*ln_GM (log difference of GM monthly stock price)
In order to calculate the return of GM, we need to have log difference of monthly GM
stock price. Use GRETL, we can compute that log difference.
Then we add variable (define new variable) with equation (1) and (2)
2. Choose Model, then Ordinary Least Squares
Choose Exr_GM as dependent variable, const and ersandp as regressors
We will have the following result:
Model 1: OLS, using observations 2002:02-2007:04 (T = 63)
Dependent variable: Exr_GM

const
ersandp
Mean dependent var
Sum squared resid
R-squared
F(1, 61)
Log-likelihood
Schwarz criterion
rho

Coefficient
0.850664
1.1488

Std. Error
1.19584
0.339127

0.604042
5475.230
0.158333
11.47520
230.0361
468.3584
0.075881

t-ratio
-0.7114
3.3875

S.D. dependent var


S.E. of regression
Adjusted R-squared
P-value(F)
Akaike criterion
Hannan-Quinn
Durbin-Watson

p-value
0.47958
0.00124

***

10.24319
9.474063
0.144535
0.001240
464.0721
465.7579
1.836330

The regresstion model, we shall have:

3. Interprete the estimates:


If the excess rate of the market portfolio increases by 1% point, the excess rate on the
GM stock increases by 1.1488 % point. And this property is true only if the regression
model contains an intercept = -0.85.
4. Compute the sample average of the residuals:
Mean of residuals = -6.69658e-016

Summary Statistics, using the observations 2002:01 - 2007:04


for the variable Residual (63 valid observations)
Mean
Median
Minimum
Maximum
-6.69658e-016
0.575088
-24.3178
21.7750
Std. Dev.
C.V.
Skewness
Ex. kurtosis
9.39735
1.40330e+016
-0.177825
0.404283
5% Perc.
95% Perc.
IQ range
Missing obs.
-17.6694
17.0471
9.43027
1
5. From result above,

= 0.158

Interpretation: 15.8% of the variation in the excess return of the GM Stock can be
explained by the regression (the variation in the excess return of S&P500 index)
Low R squared means the excess return of the GM stock doesnt act much like the
movement of the excess return of S&P500.
6. H0:
H1:
5% significance level or
We shall reject H0 if t-statistic > t(0.05,61)
t-statistic =

1.67

= 0.44

As the t-statistic < 1.67 Ho cannot be rejected on a 5% significance level in an upper


tailed test.
7. H0:
H1:
5% significance level or
We shall reject H0 if t-statistic < -t(0.05,61) -1.67
t-statistic = -0.7114 (from OLS) ,
As the t-statistic > -1.67 Ho cannot be rejected on a 5% significance level in an lower
tailed test.
8. Compute 95% Confidence Interval for
< <[

[
[0.471;1.827] is 95% Confidence Interval for the true
9. Test joint Hypothesis
H0:
and
H1:
and/or
On 5% significance level a=0.05
From GRETL, F-test(2;61) = 0.331 with p-value = 0.719
As p a (0.719 0.05) Cannot reject H0 on 5% significance level.

Exercise 2.

You might also like