checking that whether the results are economic significance and reasonable. The
estimated model is passed through diagnostic tests of the econometrics to check
its consistency and model is checked the stability of the model. Result of
Table 6 shows the Diagnostic tests: serial correlation, functional form,
normality, heteroscedasticity, and structural stability.
is found that there is no correlation i.e., The critical values for the test
are dL = 0.88, dU = 1.61, so 4 ? dU = 3.12 and 4 ? dL = 2.39. As the test
statistic (1.870818) lies between the upper (1.61) and 4 ? dU = 3.12, therefore,
hence the null hypothesis of no autocorrelation is not rejected and it would be
concluded that the residuals from the model appear to be not autocorrelated.
Therefore, the analysis satisfies the assumption of independent of errors. The model
of functional form is matched with test of Ramsey’s RESET.
As Bera-Jarque statistic p-value is 0.72 which
is greater than 0.05, therefore, do not reject the null hypothesis and it is
concluded that the residual values are normally distributed.
P-value for the F-statistic value which is 0.58 exceeds 0.05 level of
significance, therefore, it is concluded that there is no heteroscedasticity in
the variance of the residuals. In the presence of heteroscedasticity, a
remedial measure can be taken to rectify this situation by using the robust
standard errors option for correcting the (conventional) standard errors.
as all Diagnostic test have model passes all of the reported diagnostic tests. it can be say that results of study are economically
significant and reasonable because the estimated model of study have passed all
the diagnostic tests.
5.5. Test For Structure Stabiltiy Of
the help of cumulative sum of recursive residuals (CUSUM) and cumulative sum of
squares of recursive residuals (CUSUMSQ) structural stability of model are checked.
The concept of stability testing technique was first presented by Brown et al.
(1975), the stability of model of error correction model selected by ARDL can
be test with CUSUM and CUSUMSQ plots which are shown in Figures 5.1 and 5.2
respectively. According to both the plots, the plots are within critical bounds
at 5 percent level of significance, Therefore, they confirm the long-run
relationships among variables and the stability of the model.