BAYESIAN COMPARISON OF DISCRETE DISTRIBUTIONS USING MCMC
Ioannis Ntzoufras, Athanassios Katsis · 2004
The aim of this paper is to construct Bayesian model comparison tests between discrete distributions used for claim count modeling in the actuarial field. We use advanced compu-tational techniques to estimate the posterior model odds amongst different distributions for claim counts. We construct flexible reversible jump Markov Chain Monte Carlo algorithms and implement them in various illustrated examples. 1.