Victor De Oliveira
The University of Texas at San Antonio

Bayesian model selection in spatial lattice models

This work describes a Bayesian approach for model selection in Gaussian conditional autoregressive models and Gaussian simultaneous autoregressive models which are commonly used to describe spatial lattice data. The approach is aimed at situations when all competing models have the same mean structure, but differ on some aspects of their covariance structures. The proposed approach uses as selection criterion the posterior model probabilities computed using some default priors for the model parameters. The proposed methodology is illustrated using two real datasets, one dealing with phosphate concentrations on an archaeological region and the other dealing with homicide rates in southern US counties. This is joint work with J. Song.