Saint mary’s university

Testing assumptions of classical linear regression model (CLRM)

Test for absence of autocorrelation assumption (cov (ui , uj) = 0 for i _= j)

Assumption that is made of the CLRM about the disturbance terms is that the covariance between the error terms over time (or cross-sectional, for that type of data) is zero. In other words, it is assumed that the errors are uncorrelated with one another. If the errors are not correlated with one another, it would be stated that they are auto correlated or that they are serially correlated. A test of this assumption is therefore required.

To test if there is autocorrelation Q-statistics test is employed on the residual of the regression. In the statistics employed both autocorrelation and partial correlation is computed and the result found no sign of serial autocorrelation in the regression none of the probability at all lag is significant there for we cannot reject hetroscdacity of the null hypothesis.

Table 4.3 Autocorrelation

Date: 12/27/16 Time: 10:48
Sample: 2001 2014
Included observations: 84