Model Criticism and Robustness Analysis

SUMMARY This paper exposits and develops Bayesian methods of model criticism and robustness analysis. The objectives are to clarify the Bayesian interpretation of non-Bayesian diagnostic tests, and provide explicitly Bayesian procedures accessible to practical investigators. Speciic methods for prior density criticism and robustness analysis, and data density criticism, are presented. All are based on the approximationof appropriate Bayes factors, and avoid the need for posterior simulation under alternative model speciications. A general method of data density criticism is developed, that requires neither posterior simulation nor analytical approximations under any model speciication. Some of the methods presented here have been implemented in user oriented software. The paper presents a few simple illustrations of the methods.