INTEGRABLE EXPANSIONS FOR POSTERIOR DISTRIBUTIONS FOR ONE-PARAMETER EXPONENTIAL-FAMILIES

被引：0

作者：

WOODROOFE, M ^{[1
]}

机构：

[1] UNIV MICHIGAN,DEPT STAT,ANN ARBOR,MI 48109

来源：

STATISTICA SINICA | 1992年 / 2卷 / 01期

关键词：

POSTERIOR DISTRIBUTIONS; PARAMETER TRANSFORMATIONS; STEINS IDENTITY; MARTINGALE CONVERGENCE THEOREM; STOPPING TIMES; SEQUENTIAL CONFIDENCE INTERVALS;

D O I：

暂无

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：

The main results provide asymptotic expansions for posterior distributions which may be integrated termwise with respect to the marginal distribution of the data. The proof uses a data dependent transformation which converts the likelihood function to exact normality and then applies a version of Stein's identity to the posterior distributions. Applications to sequential confidence intervals are described briefly.

引用

页码：91 / 111

页数：21

共 11 条

[1] LARGE-SAMPLE THEORY OF SEQUENTIAL ESTIMATION [J].

ANSCOMBE, FJ .

PROCEEDINGS OF THE CAMBRIDGE PHILOSOPHICAL SOCIETY, 1952, 48 (04) :600-607

[2]

Berger J. O., 1984, LIKELIHOOD PRINCIPLE, V6

[3]

BICKEL PJ, 1985, P BERKELEY C HONOR J, V2, P749

[4]

BROWN LD, 1986, FUNDAMENTALS STATIST

[5]

Chow YS, 1971, GREAT EXPECTATIONS T

[6]

GHOSH JK, 1985, 3RD P PURD S STAT DE, V1, P403

[7] ASYMPTOTIC EXPANSIONS ASSOCIATED WITH POSTERIOR DISTRIBUTIONS [J].

JOHNSON, RA .

ANNALS OF MATHEMATICAL STATISTICS, 1970, 41 (03) :851-&

[8]

STEIN C, 1987, APPROXIMATE COMPUTAS

[9] VERY WEAK EXPANSIONS FOR SEQUENTIALLY DESIGNED EXPERIMENTS - LINEAR-MODELS [J].

WOODROOFE, M .

ANNALS OF STATISTICS, 1989, 17 (03) :1087-1102

[10] VERY WEAK EXPANSIONS FOR SEQUENTIAL CONFIDENCE LEVELS [J].

WOODROOFE, M .

ANNALS OF STATISTICS, 1986, 14 (03) :1049-1067

← 1 2 →