Sequential iterated bootstrap confidence intervals
成果类型：
Article
署名作者：
Lee, SMS; Young, GA
署名单位：
University of Cambridge; University of Hong Kong
刊物名称：
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
ISSN/ISSBN：
1369-7412
发表日期：
1996
页码：
235-251
关键词：
摘要：
A version of the sequential probability ratio test for testing simultaneously a set of nested hypotheses is developed. This procedure is then applied to define a sequential procedure of sampling at the inner level of the two nested levels of resampling required by Monte Carlo construction of an iterated bootstrap percentile method confidence interval. The sequential resampling scheme reduces very significantly the computational demands of construction of the iterated bootstrap confidence interval. The scheme may be applied, simply and without any adaptation, to construct a confidence interval for any parameter of interest. The performance of the sequential iterated bootstrap confidence interval is illustrated on two examples involving the ratio of two population means and a population variance.