409Hw14ans - STAT 409 Fall 2009 Homework #14 (due...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
Background image of page 9
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: STAT 409 Fall 2009 Homework #14 (due Wednesday, December 9, by 4:30 p.m.) 1. When correctly adjusted, a machine that makes widgets operates with a 5% defective b) rate. However, there is a 10% chance that a disgruntled employee kicks the machine, in which case the defective rate jumps up to 30%. Suppose that a widget made by this machine is selected at random and is found to be defective. What is the probability that the machine had been kicked? P(D) = 0.90 x 0.05 + 0.10 x 0.30 = 0.075. 0.10x 0.30 0.030 A random sample of 20 widgets was examined, 4 widgets out of these 20 are found to be defective. What is the probability that the machine had been kicked? P(X=4|K') = [240](0.05)4(0.95)16 = 0.0133, P(X=4IK) = [240)(0.30)4(0.70)16 = 0.1304. 0.10x 0.1304 2. 7.2-1 of the observations of a random sample from a Poisson distribution 7.2-1 Let Y be the sum be a gamma one with parameters at and 13. with mean 6. Let the prior p.d.f. of 6 (3) Find the posterior p.d.f. of 6, given Y = y. (b) If the loss function is [w()=) — 612, find the Bayesian point estimate w(y). (c) Show that this w(y) is a weighted average of the maximum likelihood estimate y/Il and the prior mean m8, with respective weights of 11/01 +1/f3) and (1/fi)/(n +1/fl)- (0K3 X\ X1 H X“ we. 1.1.05 POC350H __.7 1-, Z X; Cs ?o§“>sw ._ t i“ “We bxan 8f“) TV(’A\ \HBWA w at” “L r e, ‘6‘ UA+J~ *\ v? +“\ :7 Cs Gama/tax Witt“ NW 4.: fi'ho’» NW (5; t 4;; Ma (‘03 cl .LJHB : 3‘“ n +OL '56 h w 9*h “+35 (5 “+46 3. Consider a random sample X1, X2, , X” from a distribution with p.d.f. f(x;?\.)=7txk_1, O<x<l, zero elsewhere, 7V>0. A a) Find the maximum likelihood estimator of X, )L. b) Let 7x. have a prior p.d.f. which is gamma with parameters 0L and 9. Find the conditional mean of 7» given X1=x1, X2 =x2, ,Xn :26". Show that it is A a weighted average of the maximum likelihood estimator k and the prior mean 0L9. [Hint f(x;?») = ix“ = iiexh‘xq h, __ n n FA'\ m Lm —: n a 14‘: a (mi J”- ”9/ "" ":54 (‘33 fitflx @(okwtda) : Min—q \e 9x ntfiaflm‘ vme W" P (Ah-POL'l —’A(lé~~=_22~rr\) “ v . . Q, :9 \KS (1ka bu‘M,‘ NQM ,L -:_ n+9.) NW 6: k 4. Let X 1 , X 2, , X n be a random sample from the distribution with probability density function 3 f(x)=39xze_ex x>0 9>O. a) Find a sufficient statistic Y = u(X1, X2, ,Xn) for 9. f(x1,x2,...xn;6) =f(x1;9)f(x2;6) ...f(xn;9) _ez.n x.3 n _ n n F11 2 l: n By Factorization Theorem, Y = Z X is a sufficient statistic for 6. i=1 OR f(x;?») = exp{—G-x3+ln0+ln3+21nx}. : K(x)=x3. n :> Y = Z X; is a sufficient statistic for )L. i=1 A b) Obtain the maximum likelihood estimator of 9, 9 . n —Gxg L(6)=H 39x36 i=1 n n lnL(6)= n-ln9+z 1n(3xl-2)—G-Z 36-3 . _ 1 1:1 1:1 I n n " (lnL(6))=6—in3=0 :> 6: n 1:1 b) Let X1,X2,.. density function . , X n be a random sample from the distribution with probability f(x)=39xze"9x3 x>0 9>0. 12 Let W = Z X? . Find the probability distribution of W. i=1 3 FX(x) = 1 — e—ex , x>0. Let V=X3. Fv(v)=P(vsv)=P(stl/3)=1—e“9", v>0. V has an Exponential distribution with mean n n :> W = 2V2. = has aGamma distribution with 0t =11 = 12 i=1 i 1 ll .1. 9. and B: usual 9 = Use part (a) to suggest a confidence interval for 9 With ( l — Ct) 100 % confidence level. n 2 W/B = 2 9 ZX has a chi-square distribution with r = 2 OL = 2 n d.f. i=1 2 2 X _ (n) X (n) :> 1 05/2 a/2 ) : 1_a n_ n 22X? 22x? i=1 i=1 2 2 X _ (n) X (n) A ( l ~0t) 100 % confidence interval for 9: —IHL, 0::2 6. a) 7.2-4 7.2-4 Consider a random sample X1. X2 , . , . , X” from a distribution with, p.d.f. ft); £9} = flags—8"), 0 -: x <: 00. Let {9 have a prior p.d.fi which is gamma with a z 4 and the usual :9 = _1_ /4. Find the conditional mean off}, given X1 = x1, X2 x2, , X” :1» =x is b) Show that the conditional mean of 9 given X1 = x1 , X2 = x2, , X n n A a weighted average of the maximum likelihood estimator 9 and the prior mean. (What are the weights?) (M HR LH 4H3 y“ 1 "\T(e)mC(aLa,+o\\o\ :: 9 q/ i Deena1 (Ht) ‘3‘ n+LK ~\ ~9i‘iiéx33 : b o u g Q, Let X1, X2 , , X 12 be a random sample of size n = 12 from the distribution with probability density function f(x).——39x26‘ex3 x>0 G>0. Find a uniformly most powerful rejection region of size OL = 0.05 for testing H0:9=3 vs. H1:9<3. Let 9<3 n —3x3 ( ) Higxie L H -x x ...x 1:1 7» a , a = 0’ 19 2: 9 n = (XI xz x”) L(H1;X1,xz,..,xn) n 2 ~6xl3 H 39x1. 6 i=1 = {ijnexp (6-—3)§1x3 \6 i=1 1 n 7»(x1,x2,..,xn)sk C} (9—3)ZX?Sk1 n RejectHO if ac. i=1 1 n 2X? has a Gamma distribution with OL = n = 12 and B = usual 9 = i=1 n n Then -2— =’ 2 9 has a X2 ( 2 OL = 24 degrees of freedom) distribution. i=1 i=1 n 0.05 = OL = P(RejectH0|H0istrue) = P(2X§ 20 19:3) i=1 = P(6%X§’ 26c[9=3) = P(x2(24)26c). i=1 :> 60 = X30504) = 36.42. :> 0:6.07. 12 RejectHO if 26.07. i=1 ...
View Full Document

This note was uploaded on 10/12/2011 for the course STATISTICS stat 410 taught by Professor Stepanov during the Spring '11 term at University of Illinois, Urbana Champaign.

Page1 / 9

409Hw14ans - STAT 409 Fall 2009 Homework #14 (due...

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online