This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Stat 351 Fall 2007 Assignment #9 Solutions 1. (a) By Definition I, we see that X 1 ρX 2 is normally distributed with mean E ( X 1 ρX 2 ) = E ( X 1 ) ρE ( X 2 ) = 0 and variance var( X 1 ρX 2 ) = var( X 1 ) + ρ 2 var( X 2 ) 2 ρ cov( X 1 ,X 2 ) = 1 + ρ 2 2 ρ 2 = 1 ρ 2 . That is, X 1 ρX 2 = Y where Y ∈ N (0 , 1 ρ 2 ). Hence, Y = p 1 ρ 2 Z where Z ∈ N (0 , 1). In other words, there exists a Z ∈ N (0 , 1) such that X 1 ρX 2 = p 1 ρ 2 Z. 1. (b) Since X = ( X 1 ,X 2 ) is MVN, and since Z = X 1 p 1 ρ 2 ρX 2 p 1 ρ 2 , we conclude that ( Z,X 2 ) is also a MVN. Hence, we know from Theorem V.7.1 that the components of a MVN are independent if and only if they are uncorrelated. We find cov( Z,X 2 ) = cov X 1 p 1 ρ 2 ρX 2 p 1 ρ 2 ,X 2 ! = 1 p 1 ρ 2 cov( X 1 ,X 2 ) ρ p 1 ρ 2 var( X 2 ) = ρ p 1 ρ 2 ρ p 1 ρ 2 = 0 which verifies that Z and X 2 are, in fact, independent. Exercise 4.2, page 127 : If φ ( t,u ) = exp { it 2 t 2 u 2 tu } = exp { it 1 2 (4 t 2 + 2 tu + 2 u 2 ) } , then we recognize this as the characteristic function of a normal random variable...
View
Full
Document
This note was uploaded on 03/26/2012 for the course STAT 351 taught by Professor Michaelkozdron during the Fall '08 term at University of Regina.
 Fall '08
 MichaelKozdron
 Probability, Variance

Click to edit the document details