ON THE CONSISTENCY OF POSTERIOR MIXTURES AND ITS APPLICATIONS

被引:15
作者
DATTA, S
机构
关键词
POSTERIOR; CONSISTENCY; MIXING DISTRIBUTION; EMPIRICAL BAYES; ASYMPTOTIC OPTIMALITY;
D O I
10.1214/aos/1176347986
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
Consider i.i.d. pairs (theta-i, X-i), i greater-than-or-equal-to l, where theta-1 has an unknown prior distribution omega and given theta-1, X-1 has distribution P theta-1. This setup aries naturally in the empirical Bayes problems. We put a probability (a hyperprior) on the space of all possible omega and consider the posterior mean omega of omega. We show that, under reasonable conditions, P omega = integral-P-theta d omega is consistent in L1. Under a identifiability assumption, this result implies that omega is consistent in probability. As another application of the L1 consistency, we consider a general empirical Bayes problem with compact state space. We prove that the Bayes empirical Bayes rules are asymptotically optimal.
引用
收藏
页码:338 / 353
页数:16
相关论文
共 19 条
[1]  
BALDER EJ, 1983, STATIST DECISIONS, V1, P503
[2]  
Berger J.O., 1985, STAT DECISION THEORY, P74
[3]  
Billingsley P, 1968, CONVERGENCE PROBABIL
[4]   ESTIMATION OF A MIXING DISTRIBUTION FUNCTION [J].
BLUM, JR ;
SUSARLA, V .
ANNALS OF PROBABILITY, 1977, 5 (02) :200-209
[5]  
DATTA S, 1988, THESIS MICHIGAN STAT
[6]  
DUNFORD N, 1957, LINEAR OPERATOR 1
[7]  
Fabian V., 1985, INTRO PROBABILITY MA
[8]   BAYESIAN ANALYSIS OF SOME NONPARAMETRIC PROBLEMS [J].
FERGUSON, TS .
ANNALS OF STATISTICS, 1973, 1 (02) :209-230
[9]   ASYMPTOTIC SOLUTIONS TO 2 STATE COMPONENT COMPOUND DECISION PROBLEM, BAYES VERSUS DIFFUSE PRIORS ON PROPORTIONS [J].
GILLILAND, DC ;
HANNAN, J ;
HUANG, JS .
ANNALS OF STATISTICS, 1976, 4 (06) :1101-1112
[10]  
GILLILAND DC, 1982, ANN STAT, V10, P1277, DOI 10.1214/aos/1176345993