This page is hosted on AFS file server space, which is being shut down on November 13, 2018. If you are seeing this message, your service provider needs to take steps now. Visit afs.unc.edu for more information.

Lecture 38—Wednesday, April 18, 2007

What was covered?

Bayesian statistics—why should a frequentist care?

The Gibbs sampler

These are readily obtainable in hierarchical models when viewed from a Bayesian perspective. The basic regression model is a conditional statement, random effects are conditional on the values of parameters of a normal distribution, and all the remaining parameters are conditional on their priors.

,

i.e., the probability of going from  to  in n steps in the limit depends only on the final state. When a Gibbs sampler is formulated from the likelihood and priors from a Bayesian model, the limiting probability  is in fact the desired posterior probability.

The BUGS program

Course Home Page


Jack Weiss
Phone: (919) 962-5930
E-Mail: jack_weiss@unc.edu
Address: Curriculum in Ecology, Box 3275, University of North Carolina, Chapel Hill, 27516
Copyright © 2007
Last Revised--April 19, 2007
URL: http://www.unc.edu/courses/2007spring/enst/562/001/docs/lectures/lecture38.htm