There was a linear regression question that provided a scattered plot with heteroskedasticity. I ruled out 2 answer choices and was left with BLUE and asymptotic normality. Because of the heteroskedasticity I ruled out BLUE as well, but I don't remember seeing this concept of asymptotic normality in any of the review notes I had gone through. Was this concept mentioned anywhere in the notes?
Stock, Chapter 5: Single Regression: Hypothesis Tests and Confidence Intervals study notes chapter end 5th question in bionicturtle.
Both these answers in the question are true (question asks for the wrong one):
a)Whether the errors are homo- or heteroskedastic, the OLS estimators are are unbiased, consistent and asymptotically normal
c)If it is true that, in addition to the three assumptions above, that the errors are homoskedastic, then our OLS estimator for B(1) is BLUE
By the way, I don't remember the question in frm exam exactly but I remember the answer that the coefficients are asymptotically normal (A in frm question). If there is heteroscedasticity in the model then the OLS estimators are not efficient (do not have least variance) therefore I think they are not BLUE.
Stay connected