Catching up faster by switching sooner: a predictive approach to adaptive estimation with an application to the AIC-BIC dilemma

• 成果类型: Article
• 署名作者: van Erven, Tim; Grunwald, Peter; de Rooij, Steven
• 署名单位: University of Cambridge
• 刊物名称: JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
• ISSN/ISSBN: 1369-7412
• DOI: 10.1111/j.1467-9868.2011.01025.x
• 发表日期: 2012
• 页码: 361-417
• 关键词: model selection; Cross-validation; information criteria; probability; asymptotics; Consistency; complexity; PRINCIPLE; inference; entropy

摘要

. Prediction and estimation based on Bayesian model selection and model averaging, and derived methods such as the Bayesian information criterion BIC, do not always converge at the fastest possible rate. We identify the catch-up phenomenon as a novel explanation for the slow convergence of Bayesian methods, which inspires a modification of the Bayesian predictive distribution, called the switch distribution. When used as an adaptive estimator, the switch distribution does achieve optimal cumulative risk convergence rates in non-parametric density estimation and Gaussian regression problems. We show that the minimax cumulative risk is obtained under very weak conditions and without knowledge of the underlying degree of smoothness. Unlike other adaptive model selection procedures such as the Akaike information criterion AIC and leave-one-out cross-validation, BIC and Bayes factor model selection are typically statistically consistent. We show that this property is retained by the switch distribution, which thus solves the AICBIC dilemma for cumulative risk. The switch distribution has an efficient implementation. We compare its performance with AIC, BIC and Bayesian model selection and averaging on a regression problem with simulated data.