Affiliations: Statistical Laboratory, Centre for Mathematical Sciences, Cambridge, UK, and Bank of America Merrill Lynch. E-mail: gob20@statslab.cam.ac.uk | The Institute of Mathematics, Actuarial and Computer Sciences, Kingston, Jamaica. E-mail: winb365@hotmail.com
Note: [] Address for correspondence: G.O. Brown, Statistical Laboratory, Centre for Mathematical Sciences, Cambridge CB3 0WB, UK. E-mail: gob20@statslab.cam.ac.uk
Abstract: We review and apply several approaches to model selection for analysis of variance models which are used in a credibility and insurance context. The reversible jump algorithm is employed for model selection, where posterior model probabilities are computed. We then apply this method to insurance data from workers' compensation insurance schemes. The reversible jump results are compared with the Deviance Information Criterion, and are shown to be consistent.
Keywords: Reversible jump MCMC, loss ratios, Bayesian analysis, model selection