On a class of purely sequential procedures with applications to estimation and ranking and selection problems

被引:0
|
作者
Joshi, Neeraj [1 ]
Bapat, Sudeep R. [2 ]
机构
[1] Univ Delhi, Dept Stat, Delhi 110007, India
[2] Indian Inst Management, Dept Operat Management & Quantitat Tech, Indore, India
关键词
Asymptotic consistency; asymptotic efficiency; first-order asymptotics; minimum risk; multiple comparison; point estimation; purely sequential; ranking and selection; regret; second-order asymptotics; RISK POINT ESTIMATION; ESTIMATION MRPE; CONFIDENCE-INTERVALS; BOUNDED-RISK; NORMAL POPULATION; RENEWAL THEORY; PARAMETER; 2-STAGE; APPROXIMATIONS; COST;
D O I
10.1080/07474946.2021.2010407
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
In this article, we develop a general class of purely sequential procedures and obtain the associated first- and second-order asymptotics for the expected sample size and regret. We establish that many estimation and ranking and selection problems can be handled with the help of the proposed class of sequential procedures. A brief simulation analysis is carried out in support of the accuracy of our proposed sequential methodology and a real data set from environment study is included to demonstrate the practical utility.
引用
收藏
页码:482 / 500
页数:19
相关论文
共 50 条