Consequences & Detection of Heteroscedasticity

 

Consequences & Detection

of

Heteroscedasticity

Consider the SLRM

1.                    The OLS estimates are unbiased estimates in the presence of heteroscedasticity.

2.                    The OLS estimates are not efficient (minimum variance) in case of heteroscedastic disturbance term.

3.                    The OLS estimator is consistent in the presence of heteroscedasticity.

The variance of OLS estimates is 

4.      In the presence of heteroscedasticity, student’s two samples t - test and ANOVA are not applicable. In two samples t – test and ANOVA it is assumed that the population variances should be identical.

4.      The Gauss Markov’s theorem assures that the OLS estimator minimum variance than all other unbiased estimators; nevertheless, in the presence of heteroscedasticity, the OLS estimators are no longer BLUE.