ABSTRACT
Attainment of convergence of McMC algorithms does not necessarily yield good models. If a model is mis-specified then it will be of limited use. There are many issues relating to model goodness-of-fit that should be of concern when evaluating models for geo-referenced disease data. In this section, I treat general issues related to the use of goodness-of-fit (GOF) measures, residual diagnostics and the use of posterior output to yield risk exceedence probabilities.
