[S] summary: maximization; MCMC; random number

jacquier@jacquier.bc.edu
Mon, 11 May 1998 14:20:46 -0400 (EDT)


2. MCMC sampling
The followings are found to be of concern in sampling
a. acceptance rate; adjusting the proposal distribution variance to have a
acceptance rate at around 30% (don't know why, but it converges faster)

How can there be a fit all approach to this problem . Let's us assume that
I am trying to cover a normal with a normal. So I can do it perfectly and
I should shoot for an acceptance rate of 100%. Clearly the acceptance
rate is a function of the shape difference between the cover and the
distribution you want to draw from. Am I missing something here?
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news