An Optimal Confidence Region for the Largest and the Smallest Means from a Multivariate Normal Distribution

淡江大學機構典藏 > 商管學院 > 統計學系暨研究所 > 期刊論文 > Item 987654321/108152

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/108152

题名:	An Optimal Confidence Region for the Largest and the Smallest Means from a Multivariate Normal Distribution
作者:	Wu, Shu-Fei;Chen, Hubert J.
关键词:	Confidence region;Fixed-width;Largest (smallest) mean;Least favorable configuration;Student
日期:	2016-08-03
上传时间:	2016-10-25 02:10:30 (UTC+8)
出版者:	Taylor & Francis Inc.
摘要:	An optimal confidence region is proposed for obtaining the largest and the smallest means of a multivariate normal distribution with a common unknown variance and a non-negative correlation coefficient. The confidence region is compared to a class of asymmetric confidence regions, and the results show that it is uniformly better. Further, a design-oriented two-stage confidence region with a fixed width is also given. Finally, the optimal confidence region is applied to an experiment to measure the treatment effectiveness of a physical therapy with four independent groups and the result reveals that the proposed confidence region can provide some useful information.
關聯:	Communications in Statistics: Simulation and Computation 45(3), pp.952-967
DOI:	10.1080/03610918.2014.889160
显示于类别:	[統計學系暨研究所] 期刊論文

文件中的档案:

档案	描述	大小	格式	浏览次数
An Optimal Confidence Region for the Largest and the Smallest Means from a Multivariate Normal Distribution.pdf		370Kb	Adobe PDF	0	检视/开启
index.html		0Kb	HTML	206	检视/开启

在機構典藏中所有的数据项都受到原著作权保护.

TAIR相关文章