Applied Econometrics for Accounting and Finance
Assignment # 2 – May 2023
1. a) Fit a simple linear regression model of asset price on Standard & Poor’s market price index.
The fitted model of asset price (general electric) on Standard & Poor’s (S&P 500) stock
market index shown below.
Dependent Variable: GENERAL_ELECTRIC
Method: Least Squares
Date: 06/04/23 Time: 13:56
Sample: 2002M01 2013M04
Included observations: 136
Variable Coefficient Std. Error t-Statistic Prob.
C 0.227937 2.422976 0.094073 0.9252
S_AND_P 0.017756 0.001998 8.888346 0.0000
R-squared 0.370900 Mean dependent var 21.48925
Adjusted R-squared 0.366205 S.D. dependent var 5.653229
S.E. of regression 4.500606 Akaike info criterion 5.860898
Sum squared resid 2714.230 Schwarz criterion 5.903731
Log likelihood -396.5410 Hannan-Quinn criter. 5.878304
F-statistic 79.00270 Durbin-Watson stat 0.053560
Prob(F-statistic) 0.000000
1b) Plot the residuals against time. Do you suspect the presence of residual autocorrelation?
Provide a brief discussion.
Auto correlation diagnostic in E view
Graphical method
1c).Test for the presence of serial correlation using the Breusch-Godfrey Serial Correlation
LM Test with two lags (the default in EViews). Also plot the partial autocorrelation function
(PACF) of the OLS residuals and determine the order of autocorrelation.
Breusch-Godfrey Serial Correlation LM Test:
F-statistic 976.5874 Prob. F(2,132) 0.0000
Obs*R-squared 127.3906 Prob. Chi-Square(2) 0.0000
Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 06/04/23 Time: 14:38
Sample: 2002M01 2013M04
Included observations: 136
Presample missing value lagged residuals set to zero.
Variable Coefficient Std. Error t-Statistic Prob.
C -0.148742 0.615829 -0.241531 0.8095
S_AND_P 8.76E-05 0.000508 0.172535 0.8633
RESID(-1) 1.031310 0.087302 11.81310 0.0000
RESID(-2) -0.058218 0.087846 -0.662727 0.5087
R-squared 0.936696 Mean dependent var -1.12E-15
Adjusted R-squared 0.935257 S.D. dependent var 4.483906
S.E. of regression 1.140912 Akaike info criterion 3.130503
Sum squared resid 171.8218 Schwarz criterion 3.216170
Log likelihood -208.8742 Hannan-Quinn criter. 3.165316
F-statistic 651.0583 Durbin-Watson stat 1.802558
Prob(F-statistic) 0.000000
The PACF OF OLS residual shown below
Correlgram of residual
1d) If the disturbances are auto correlated, apply the Cochrane-Orcutt transformation and re-estimate
the model. Has the transformation solved the problem?
Correcting for auto correlation
1.1 Result of the fitted modal an estimated of P
Residual residual (-1)
Dependent Variable: RESIDUAL
Method: Least Squares
Date: 06/04/23 Time: 15:41
Sample (adjusted): 2002M02 2013M04
Included observations: 135 after adjustments
Variable Coefficient Std. Error t-Statistic Prob.
RESIDUAL(-1) 0.975197 0.020034 48.67683 0.0000
R-squared 0.946469 Mean dependent var -0.039902
Adjusted R-squared 0.946469 S.D. dependent var 4.476304
S.E. of regression 1.035670 Akaike info criterion 2.915354
Sum squared resid 143.7300 Schwarz criterion 2.936874
Log likelihood -195.7864 Hannan-Quinn criter. 2.924099
Durbin-Watson stat 1.968457
1.2 Eviews output of OLS regeration of general Electric on S and p transformation is shown
below
Dependent Variable: GENERAL_ELECTRIC TRANS
Method: Least Squares
Date: 06/04/23 Time: 16:11
Sample (adjusted): 2002M02 2013M04
Included observations: 135 after adjustments
Variable Coefficient Std. Error t-Statistic Prob.
C -0.129025 0.107142 -1.204244 0.2306
S_AND_P TRANS 0.019214 0.001813 10.59952 0.0000
R-squared 0.457917 Mean dependent var 0.504220
Adjusted R-squared 0.453841 S.D. dependent var 1.398294
S.E. of regression 1.033375 Akaike info criterion 2.918241
Sum squared resid 142.0259 Schwarz criterion 2.961283
Log likelihood -194.9813 Hannan-Quinn criter. 2.935732
F-statistic 112.3498 Durbin-Watson stat 1.999898
Prob(F-statistic) 0.000000
1.3 the PACF OF the residuals in the transformed modal is shown below
1.4 . breush godfry test of error auto correlation
Breusch-Godfrey Serial Correlation LM Test:
F-statistic 1.831747 Prob. F(2,131) 0.1642
Obs*R-squared 3.672648 Prob. Chi-Square(2) 0.1594
Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 06/04/23 Time: 16:29
Sample: 2002M02 2013M04
Included observations: 135
Presample missing value lagged residuals set to zero.
Variable Coefficient Std. Error t-Statistic Prob.
C -0.001647 0.106811 -0.015424 0.9877
S_AND_P1 0.000165 0.001823 0.090730 0.9278
RESID(-1) -0.023697 0.087514 -0.270775 0.7870
RESID(-2) -0.167074 0.087783 -1.903259 0.0592
R-squared 0.027205 Mean dependent var 9.46E-17
Adjusted R-squared 0.004927 S.D. dependent var 1.029512
S.E. of regression 1.026973 Akaike info criterion 2.920289
Sum squared resid 138.1621 Schwarz criterion 3.006372
Log likelihood -193.1195 Hannan-Quinn criter. 2.955271
F-statistic 1.221165 Durbin-Watson stat 2.051804
Prob(F-statistic) 0.304670
