Part of Advances in Neural Information Processing Systems 20 (NIPS 2007)
Tim Erven, Steven Rooij, Peter Grünwald
Bayesian model averaging, model selection and their approximations such as BIC are generally statistically consistent, but sometimes achieve slower rates of con- vergence than other methods such as AIC and leave-one-out cross-validation. On the other hand, these other methods can be inconsistent. We identify the catch-up phenomenon as a novel explanation for the slow convergence of Bayesian meth- ods. Based on this analysis we deﬁne the switch-distribution, a modiﬁcation of the Bayesian model averaging distribution. We prove that in many situations model selection and prediction based on the switch-distribution is both consistent and achieves optimal convergence rates, thereby resolving the AIC-BIC dilemma. The method is practical; we give an efﬁcient algorithm.