Question: Take a regression model Y Xe with E[e j X] 0 and i.i.d. observations (Yi ,Xi ) and scalar X. The parameter of
Take a regression model Y Æ X¯Åe with E[e j X] Æ 0 and i.i.d. observations (Yi ,Xi ) and scalar X. The parameter of interest is µ Æ ¯2. Consider the OLS estimators b¯ and bµ Æ b¯2.
(a) Find E
£bµ j X
¤
using our knowledge of E
£ b¯ j X
¤
and Vb¯
Æ var
£ b¯ j X
¤
. Is bµ biased for µ?
(b) Suggest an (approximate) biased-corrected estimator bµ¤ using an estimator b Vb¯ for Vb¯.
(c) For bµ¤ to be potentially unbiased, which estimator of Vb¯ is most appropriate?
Under which conditions is bµ¤ unbiased?
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
