Unformatted text preview: X ∗ t other than X t − V t . In e f ect, we need to & nd the part of V t that is uncorrelated with X t . To do so, we construct W t = aV t − bX ∗ t , for values of a and b to be determined. The values of a and b must be such that E ( X t W t ) = 0 and also satisfy X ∗ t = cX t − W t . Find the values of ( a,b, c ) and use the resultant W t to derive the exact & nite sample bias of the OLS estimator....
View
Full Document
 Fall '08
 Staff
 Economics, Least Squares, Normal Distribution, Regression Analysis, Þnite sample bias, exact Þnite sample

Click to edit the document details