Variance under random parameters

Let MM and SS be independent Exponential(1)\text{Exponential}(1) random variables. Given M=μM = \mu and S=σS = \sigma, the pair (X,Y)(X, Y) is bivariate normal with both means μ\mu, both variances σ2\sigma^2, and correlation ρ\rho (a fixed constant with ρ1|\rho| \le 1). Compute Var(Y)\operatorname{Var}(Y).

Show hints (2)+
  1. Law of total variance: Var(Y)=E[Var(YM,S)]+Var(E[YM,S])=E[S2]+Var(M)\operatorname{Var}(Y)=\mathbb{E}[\operatorname{Var}(Y\mid M,S)]+\operatorname{Var}(\mathbb{E}[Y\mid M,S])=\mathbb{E}[S^2]+\operatorname{Var}(M).
  2. For Exp(1)\text{Exp}(1): Var(M)=1\operatorname{Var}(M)=1, E[S2]=2\mathbb{E}[S^2]=2. (ρ\rho doesn't affect Var(Y)\operatorname{Var}(Y).)

Answer

Reveal answer →

3

Want the full step-by-step worked solution? It's part of Premium - along with a worked solution for every question in the bank.

Asked at: Citadel, Two Sigma

Related questions