A note on mean-field variational approximations in Bayesian probit models

被引：4

作者：

Armagan, Artin ^{[1
]}

Zaretzki, Russell L. ^{[2
]}

机构：

[1] Duke Univ, Dept Stat Sci, Durham, NC 27708 USA

[2] Univ Tennessee, Dept Stat Operat & Management Sci, Knoxville, TN 37996 USA

来源：

COMPUTATIONAL STATISTICS & DATA ANALYSIS | 2011年 / 55卷 / 01期

关键词：

Variational inference; Bayesian probit model; Gibbs sampling;

D O I：

10.1016/j.csda.2010.06.005

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

We correct some conclusions presented by Consonni and Mann (2007) on the performance of mean-field variational approximations to Bayesian inferences in the case of a simple probit model. We show that some of their presentations are misleading and thus their results do not fairly present the performance of such approximations in terms of point estimation under the specified model. (C) 2010 Elsevier B.V. All rights reserved.

引用

页码：641 / 643

页数：3

共 5 条

[1]

ALBERT JH, 1993, J AM STAT ASS, V88

[2] Mean-field variational approximate Bayesian inference for latent variable models [J].