Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 3

Diagnostic Test Result

1. Jarque-Bera (JB) Test


10

Series: Residuals
Sample 1985 2014
Observations 30

Mean
Median
Maximum
Minimum
Std. Dev.
Skewness
Kurtosis

2.96e-15
0.004918
0.095886
-0.110776
0.047503
-0.198100
3.043893

Jarque-Bera
Probability

0.198626
0.905459

0
-0.10

-0.05

0.00

0.05

0.10

H0 : The error term is normally distributed.


Ha: The error term is not normally distributed.
The figure above showed the normality tests result for the estimated model. Based
on the figure, the p-value of the Jarque-Bera (JB) test is larger than 5% level of
significant. According to the rejection rules, we do not reject the null hypothesis, since
the level of significant (0.05) is less than probability (0.905459) which is not statistically
significant. The estimated model does not have the normality problem. Therefore, the
error term is normally distributed.

2. Serial Correlation (Breusch- Godfrey Test)


Breusch-Godfrey Serial Correlation LM Test:
F-statistic
Obs*R-squared

1.647647
3.621826

Prob. F(2,24)
Prob. Chi-Square(2)

0.2135
0.1635

Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 12/11/15 Time: 10:58
Sample: 1985 2014
Included observations: 30
Presample missing value lagged residuals set to zero.
Variable

Coefficient

Std. Error

t-Statistic

Prob.

LEX
LIM
RER
C
RESID(-1)
RESID(-2)

0.056555
-0.052361
-0.020213
-0.053104
0.346838
0.061444

0.144215
0.141352
0.035549
0.286695
0.204977
0.232157

0.392158
-0.370428
-0.568584
-0.185228
1.692084
0.264664

0.6984
0.7143
0.5749
0.8546
0.1036
0.7935

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)

0.120728
-0.062454
0.048964
0.057539
51.27916
0.659059
0.657770

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

2.96E-15
0.047503
-3.018611
-2.738371
-2.928960
1.745776

H0 : The error term does not exhibit serial correlation problem.


Ha : The error term exhibit serial correlation problem.
The figure above showed the autocorrelation tests result for the estimated model.
Based on the figure, the p-value of the serial correlation (Breusch- Godfrey) test is larger

than 5% level of significant. According to the rejection rules, we do not reject the null
hypothesis, since the level of significant (0.05) is less than probability (0.2135) which is
not statistically significant. The estimated model does not have the autocorrelation
problem. Therefore, the error term does not exhibit serial correlation problem.

You might also like