Generalized information criteria for optimal Bayes decisions 1 Research Report No . 2239 December 2008 Generalized information criteria for optimal Bayes decisions

This report constitutes an unrefereed manuscript which is intended to be submitted for publication. Any opinions and conclusions expressed in this report are those of the author(s) and do not necessarily represent the views of the Institute. This paper deals with Bayesian models given by statistical experiments and common types of loss functions. Probability of error of the Bayes identificator of state, and of more general types of Bayes risk are characterized by means of classical and generalized information criteria applicable to the experiment. In particular, the accuracy of the approximation is studied. A number of concrete numerical results and figures illustrate the obtained theoretical results.