文章基本信息

标题：Moment Matching Priors
本地全文：下载
作者：Malay Ghosh University of Florida, Gainesville ; USA Ruitao Liu University of Iowa ; Iowa City, USA 等
期刊名称：Sankhya. Series A, mathematical statistics and probability
印刷版ISSN：0976-836X
电子版ISSN：0976-8378
出版年度：2011
卷号：73
期号：02
页码：185--201
出版社：Indian Statistical Institute
摘要：There are various proposals for the selection of the so-called \objective" or \default" priors in Bayesian analysis. The paper introduces a new criterion, the moment match- ing criterion, which requires the matching of the posterior mean with the maximum likelihood estimator up to a high order of approximation. A complete characterization of such priors in the one or multi-parameter case is provided. In the process, many new priors are derived. One interesting nding is that even in the absence of nuisance parameters, it is possible to nd priors di erent from Je reys' prior for a real valued parameter based on our criterion.
关键词：Asymptotic expansion, exponential family, location-scale family, maximum likelihood, posterior mean, proper dispersion models.