HW 4 2.3 Need clarification about the hints, do not know what to do

[Yitong Zhou]

I think what I need to prove, is N*NG=NG. But the tricky part is, for hint 1, should I write both the normal distribution in N and NG as the form of \theta and sufficient statistics? but there should be some variable changing, and I get confused about what the set of variables I am supposed to use... Especially m_0 is very confusing, since we already have the u_0, why do we need m_0? And how about the empirical variances, why we do not need that?

[Mu Li]

Since both mean and variance are unknown, you need both prior u_0 and m_0. You can check the wikipedia paper for better understanding

http://en.wikipedia.org/wiki/Conjugate_prior

On Sun, Mar 31, 2013 at 4:42 PM, Yitong Zhou <timy...@gmail.com> wrote:

I think what I need to prove, is N*NG=NG. But the tricky part is, for hint 1, should I write both the normal distribution in N and NG as the form of \theta and sufficient statistics? but there should be some variable changing, and I get confused about what the set of variables I am supposed to use... Especially m_0 is very confusing, since we already have the u_0, why do we need m_0? And how about the empirical variances, why we do not need that?

--
http://alex.smola.org/teaching/cmu2013-10-701 (course website)
http://www.youtube.com/playlist?list=PLZSO_6-bSqHQmMKwWVvYwKreGu4b4kMU9 (YouTube playlist)
---
You received this message because you are subscribed to the Google Groups "10-701 Spring 2013 CMU" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 10-701-spring-201...@googlegroups.com.
To post to this group, send email to 10-701-spri...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.