Question

Do you agree with the following statements and why?

Heteroscedasticity has never been a reason to throw out an otherwise good model

Answer 1

Solution:

Heteroskedasticity cause due to violation of OLS assumption: , that is the variance of each disturbance term condition on the chosen value of the explanatory variable is constant

Although in the case of heteroskedasticity, the coefficients remain unbiased and consistent but they are inefficient. This inefficiency, the test of hypothesis like t-test, F-test, etc,  is no longer valid. For example:

formula for t-stat:

So from this formula, we can see that if the coefficient will be inconsistent, their variance may be overestimated or underestimated and hence the hypothesis test will not give the reliable result.