A bayesian approach to sequential estimation without using the
作者:
Leng Cheng Hwang,
期刊:
Sequential Analysis
(Taylor Available online 1997)
卷期:
Volume 16,
issue 4
页码: 319-343
ISSN:0747-4946
年代: 1997
DOI:10.1080/07474949708836391
出版商: Marcel Dekker, Inc.
关键词: asymptotically Bayes;asymptotically non-deficient;asymptotically pointwise optimal;empirical Bayes
数据来源: Taylor
摘要:
The problem of Bayes sequential point estimation without using the prior information is considered. Two sequential procedures of the type of Bickel and Yahav (1967, 1968), for the scale exponential and location normal families respectively, are proposed. It is shown in the present paper that the proposed procedures are asymptotically pointwise optimal (A.P.O.) in the sense of Bickel and Yahav (1967) and the second order approximation of the Bayes risk is established for a large class of prior distributions.
点击下载:
PDF (2203KB)
返 回