Linear Combination of Two Random Variables • Let X1 and X2 be random variables with E X 1 1 , E X 2 2 , var X 1 12 , var X 2 22 and Cov X 1 , X 2 12 • Let Y = aX1 + bX2. Then,… week2 1 Linear Combination of Independent R.V • Let X1, X2,…, Xn be independent random variables with EX i i • and var X i i2 Let Y = a1X1 + a2X2 + ∙∙∙+ anXn. Then,… week2 2 Linear Combination of Normal R.V • Let X1, X2,…, Xn be independent normally distributed random variables with EX i i and var X i i2 • Let Y = a1X1 + a2X2 + ∙∙∙+ anXn. Then,… week2 3 Examples week2 4 The Chi Square distribution • If Z ~ N(0,1) then, X = Z2 has a Chi-Square distribution with parameter 1, i.e., X ~ 21 . • Can proof this using change of variable theorem for univariate random variables. • The moment generating function of X is 1/ 2 1 m X t 1 2t week2 5 Important Facts 2 2 2 • If X 1 ~ v1 , X 2 ~ v2 , , X k ~ vk , all independent then, k T X i ~ 2k v i 1 1 i • If Z1, Z2,…Zn are i.i.d N(0,1) random variables. Then, Y X 12 X 22 X n2 ~ 2n week2 6 Claim • Suppose X1, X2,…Xn are i.i.d normal random variables with mean μ and variance σ2. Then, Z i X i are independent standard normal variables, where i = 1, 2, …, n and Xi 2 Z ~ n i i 1 i 1 n 2 n week2 7 Important facts • Suppose X1, X2,…Xn are i.i.d normal random variables with mean μ and variance σ2. Then, 1 2 X n i 1 X ~ 2n 1 2 i Proof:… • This implies that… n 1S 2 2 ~ 2n 1 • Further, it can be shown that X and s2 are independent. week2 8