Slice Gibbs Sampling for Simulation Based Fitting of Spatial Data Models

An auxiliary variable method which we refer to as a slice Gibbs sampler is shown to provide an attractive simulation-based model tting strategy for tting Bayesian models under proper priors. Though broadly applicable, we illustrate in the context of tting spatial models for geo-referenced or point source data. Spatial modeling within a Bayesian framework ooers inferential advantages and the slice Gibbs sampler provides an algorithm which is essentially \oo the shelf". Further advantages over importance sampling approaches and Metropolis approaches are noted and illustrative examples are supplied.