Jun Zhu, Ph.D.
Associate Professor, Department of Statistics
University of Wisconsin, Madison
Monday., September 17, 2007
A spatial-temporal autologistic regression model relates a binary response variable to potential covariates, while accounting for both spatial and temporal correlation. This modeling framework is flexible and can be useful for analyzing spatial-temporal binary data. Here I discuss various computational techniques for statistical inference including maximum pseudolikelihood, Monte Carlo maximum likelihood, and Bayesian hierarchical modeling. These approaches are illustrated by a real data example of bark beetle outbreaks.