Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

$X$ has a normal distribution. The conditional distribution of another random variable $Y$ given that $X=x$ is a normal distribution with mean $ax+b$ and variance $t^2$, where $a$, $b$, and $t^2$ are constants. How can I prove that the joint distribution of $X$ and $Y$ is bivariate normal distribution?

share|improve this question

1 Answer 1

up vote 6 down vote accepted

By definition, $$f_{X,Y}(x,y)=f_X(x)\cdot f_{Y\mid X}(y\mid x)=\frac1{\sqrt{2\pi\sigma^2}}\mathrm e^{-(x-\mu)^2/(2\sigma^2)}\cdot\frac1{\sqrt{2\pi t^2}}\mathrm e^{-(y-ax-b)^2/(2t^2)} $$ Hence the task is to solve for $M$ in $\mathbb R^2$ and $C$ a definite positive $2\times2$ matrix, the fact that for every $z=(x,y)^t$, one has the identity $$ f_{X,Y}(z)=\frac1{2\pi\sqrt{\det C}}\mathrm e^{-\frac12(z-M)^tC^{-1}(z-M)}. $$ Hints: note that the diagonal of $C^{-1}$ is $(1/\sigma^2+a^2/t^2,1/t^2)$ and that $M=(\mu,a\mu+b)^t$. The rest of the proof relies on some simple algebraic manipulations of a degree $2$ polynomial in $(x,y)$.

share|improve this answer
    
thank you for your clue –  perry zhu Jul 6 '12 at 15:02
    
You are welcome. Why did you unaccept this answer? –  Did Jul 8 '12 at 19:45
    
Haha! +1 $ $ $ $ –  Tim Jul 8 '12 at 20:18

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.