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.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

$u$~$N(0,A)$ and $z|u$~$N(u,1)$ how to show that $u|z$~$N(Bz,B)$ where $B=A/(A+1)$ ?

Empirical Bayes and Large Scale Interference

share|cite|improve this question
Knowing $f_U$ and $f_{Z\mid U}$ yields $f_{U,Z}$, which yields $f_{U\mid Z}$. Which part of this program have you trouble with? – Did Jan 28 '13 at 16:11
what is $f(U,Z)$ ? $f(U,Z)=?$ – Qbik Jan 28 '13 at 16:26
Never wrote $f(U,Z)$, but $f_{U,Z}$ the joint density of $(U,Z)$. – Did Jan 28 '13 at 16:53
Which part of this program have you trouble with? – Did Jan 28 '13 at 16:53
Somebody is erasing their footprints... – Did Jan 30 '13 at 22:18

Note that

$$p(\mu|z) \propto p(z|\mu)p(\mu)$$

After you plug it $p(z|\mu)$ and $p(\mu)$ and you will find $p(\mu|z)$ has a pdf that is normal. And you can easily derive its mean and variance.

share|cite|improve this answer
but what is $p(z)$ ? $N(0,1)$ ? – Qbik Jan 28 '13 at 16:22
@Qbik There is no $p(z)$ involved here. You only need the conditional distribution of $p(z|\mu)$. – Patrick Li Jan 28 '13 at 16:55

Your Answer


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.