This preview shows page 1. Sign up to view the full content.
Unformatted text preview: 2 by calculating: V 2 t = 1 n n X i =1 ( X i & & ) 2 : Explain why the method suggested by Tom is infeasible. (ii) Amy, knowing the problem that V 2 t has, suggested instead to construct the estimator of 2 as follows: V 2 a = 1 n n X i =1 & X i & & X 2 : (A) Is the method suggested by Amy feasible? (B) Is V 2 a biased? If yes, what is the bias of V 2 a ? [Hint: Rewrite Amy&s estimator as V 2 a = 1 n n X i =1 & X i & & + & & & X 2 = = 1 n n X i =1 ( X i & & ) 2 & ( & X & & ) 2 and evaluate its expectation using (a)(1) and (a)(2).] (C) Show that V 2 a is asymptotically unbiased though. (D) Based on the result in (b)(2)(2), show that an (even better) unbiased estimator of 2 is given by: S 2 = 1 n & 1 n X i =1 & X i & & X 2 : (Note: S 2 is called the sample variance .) 1...
View Full
Document
 '07
 HONG

Click to edit the document details