Bayesian Model Averaging By Mixture Modeling

11/27/2017

∙

by   Merlin Keller, et al.

∙

0

∙

share

A new and numerically efficient method for Bayes factor computation and Bayesian model averaging, seen as a special case of the mixture model approach for Bayesian model selection in the seminal work of Kamari, 2014. Inheriting from the good properties of this approach, it allows to extend classical Bayesian model selection/averaging to cases where improper priors are chosen for the common parameter of the candidate models.