This preview shows page 1. Sign up to view the full content.
Unformatted text preview: PROF HOHG FALL 2007 ECONOMICS 619 PROBLEM SET # 11 1. [# 7.1, p.355] One observation in taken on a discrete random variable X with pmf f (x; ), where 2 f1; 2; 3g. Find the MLE of x 0 1 2 3 f (xj 1)
1 3 1 3 1 6 1 6 f (xj 2)
1 4 1 4 1 4 f (xj 3) 0 0
1 2 1 4 0 2. [# 7.7, p.355] Let X1( Xn be i.i.d. with one of two pdfs. If ; :::; 1 0<x<1 f (x; ) = 0 otherwise while if = 1, then ( p 1=(2 x) 0 < x < 1 f (x; ) = 0 otherwise Find the MLE of . = 0, then 3. [# 7.8, p.356] One observation, X, is taken from a N (0; 2 ) population. (a) Find an unbiased estimation of 2 : (b) Find the MLE of : (c) Discuss how the method of moments estimator of might be founded. 4. [# 7.19, p.358] Suppose that the random variables Y1 ; :::; Yn satisfy Yi = xi + "i ; i = 1; :::; n; where x1 ; :::; xn are ...xed constants, and are i.i.d. N (0; 2 ), 2 are unknown. (a) Find a twodimensional su cient statistic for ( ; 2 ): (b) Find the MLE of , and show that it is an unbiased estimator of : (c) Find the distribution of the MLE of : 5. [# 7.40, p.362] Let X1 ; :::; Xn be i.i.d. Bernoulli(p). Show that the variance of Xn attains the CramerRao Lower Bound, and hence Xn is the best unbiased estimator of p: 6. [# 7.42, p.363] Let W1 ; :::; Wk be unbiased estimators of a paramater with Var Wi = 2 and Cov(Wi ; Wj ) = 0 if i 6= j: P (a) Show that, of all estimators of the form k ai Wi ; where the ai s are constants and i=1 Pk 2 Pk i=1 Wi = i Pk E ( i=1 ai Wi ) = , the estimator W = 2 has minimum variance.
i=1 1= i 1 (b) Show that Var(W ) = 7. [# 7.44, p.363] Let X1 ; :::; Xn be i.i.d. N ( ; 1). Show that the best unbiased estimator of 2 2 is Xn (1=n): Calculate its variance and show that it is greater than the CramerRao Lower 1 2 Bound. [Hint: You can calculate the variance of Xn n using the property of 2 .] 8. [# 7.50(a,b), p.364] Let X1 ; :::; Xn be p i.i.d. N ( ; 2 ); > 0: For this model both Xn and cS 1 are unbiased estimators of , where c = n p2 ((n 1)=2) : (n=2) (a) Prove that for any number a the estimator aXn + (1 a)cSn is an unbiased estimator of . (b) Find the value of a that produces the estimator with minimum variance. 9. Suppose ^1 ; ^2 and ^3 are estimators of ; and we know that E(^1 ) = E(^2 ) = ; E(^3 ) 6= ; var(^1 ) = 12; var(^2 ) = 10; and E(^3 )2 = 6: Which estimator is the best in terms of MSE criterion? 10. Suppose fX1 ; X2 ; :::; Xn g is an i.i.d. random sample from some population with unknown mean and variance 2 : De...ne parameter = ( 2)2 : (a) Suppose ^ = (Xn 2)2 is an estimator for ; where Xn is the sample mean: Show that ^ is not unbiased for : [Hint: Xn 2 = Xn + 2:] (b) Find an unbiased estimator for : 11. A random sample, X1 ; :::; Xn ; is taken from an i.i.d. population with ( ; the following estimator of : X 2 2 ^= (X1 + 2X2 + 3X3 + i Xi = n(n + 1) n(n + 1)
i=1 n 2 ): Pk 1
i=1 1= 2 i : Consider + nXn ) : (a) Show ^ is unbiased for : Pn (b) Which estimator, ^ or Xn ; is more e cient? Explain. [Hint: i=1 i = Pn 2 n(n+1)(2n+1) :] i=1 i = 6 12. Suppose (X1 ; X2 ; :::; Xn ) is an i.i.d. N (0;
2 Sn 2) n(n+1) 2 and random sample. De...ne (Xi Xn )2 ; = (n 1) 1 where X1 = n 1 Pn n X i=1 i=1 Xi ; and ^2 = n
1 n X i=1 Xi2 : Show which estimator is more e cient? Give your reasoning. 2 ...
View
Full
Document
This homework help was uploaded on 12/08/2007 for the course ECON 6190 taught by Professor Hong during the Fall '07 term at Cornell University (Engineering School).
 Fall '07
 HONG
 Economics, Econometrics

Click to edit the document details