2 - People.stat.sfu.ca
2 - People.stat.sfu.ca
2 - People.stat.sfu.ca
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
I intended you to: find the joint density of X, Z by change of variables,<br />
and then divide joint by marginal to get the conditional densities.<br />
Here is one way to do part of the problem. Consider first the following<br />
problem: U, V are bivariate normal with mean 0 and variance covariance<br />
matrix [<br />
1<br />
]<br />
ρ<br />
ρ 1<br />
Then<br />
f U,V (u, v) =<br />
1<br />
2π √ − 2ρuv + v 2<br />
1 − ρ 2 exp{−u2 }<br />
2(1 − ρ 2 )<br />
Rewrite u 2 − 2ρuv + v 2 as (u − ρv) 2 + (1 − ρ 2 )v 2 to see that<br />
f V (v) =<br />
1<br />
2π √ 1 − ρ 2 e−v2 /2<br />
∫ ∞<br />
−∞<br />
e −(u−ρv)2 /[2(1−ρ 2 )] du<br />
In the integral substitute y = (u − ρv)/ √ 1 − ρ 2 and find that<br />
f V (v) = 1 √<br />
2π<br />
e −v2 /2<br />
The conditional density of U given V = v is then<br />
f U|V (u|v) = f U,V (u, v)<br />
f V (v)<br />
which is the N(ρv, 1 − ρ 2 ) density.<br />
√<br />
(u − ρv)2<br />
= exp{−<br />
2(1 − ρ 2 ) − v2<br />
2 + v2 2π<br />
2 } 2π<br />
= exp{−(u − ρv) 2 /[2(1 − ρ 2 )]}/ √ 2π(1 − ρ 2 )<br />
Now if X = σU + µ and Z = φV + µ + γ then X, Z has the joint<br />
distribution of X and X + Y of the original problem provided we make<br />
ρ = Corr(X, Z). For X and Z as in the problem we find φ 2 = σ 2 + τ 2 ,<br />
the covariance is σ 2 , and the correlation is<br />
ρ =<br />
1<br />
√<br />
1 + τ<br />
2<br />
/σ 2 =<br />
σ<br />
√<br />
σ2 + τ 2 = σ φ<br />
8