This preview shows page 1. Sign up to view the full content.
Unformatted text preview: n.
a. First rewrite this estimator:
( ̃ ) and then take the expectation:
[̃] [ [̃]
b. Using the previous result:
[̃] [ [ ( ) page 3 [ (
c. To show that the variance is greater than the OLS estimator note that ̃
is unbiased (shown in (a)) and that it is linear in , with weights
̃ must have a larger variance.
Since OLS is the BLUE, 8. For individuals of equal education, the predicted ( ) BLACK/WHITE gap is The coefficient on the dummy is a positive 1.7%. But that’s not very interesting because
it tells us the wage difference at age zero, when relatively few people are employed. In
View Full Document