Relevant statistics for Bayesian model choice

Autor:	Marin, J. -M., Pillai, N., Robert, C. P., Rousseau, J.
Rok vydání:	2011
Předmět:	Mathematics - Statistics Theory Quantitative Biology - Populations and Evolution Statistics - Computation Statistics - Methodology
Druh dokumentu:	Working Paper
Popis:	The choice of the summary statistics used in Bayesian inference and in particular in ABC algorithms has bearings on the validation of the resulting inference. Those statistics are nonetheless customarily used in ABC algorithms without consistency checks. We derive necessary and sufficient conditions on summary statistics for the corresponding Bayes factor to be convergent, namely to asymptotically select the true model. Those conditions, which amount to the expectations of the summary statistics to asymptotically differ under both models, are quite natural and can be exploited in ABC settings to infer whether or not a choice of summary statistics is appropriate, via a Monte Carlo validation. Comment: 30 pages, 8 figures, 1 table
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/1110.4700 Zobrazit plný text záznamu View this record from Arxiv