A test for the presence of stochastic ordering under censoring: the k-sample case

被引:0
|
作者
El Barmi, Hammou [1 ]
机构
[1] CUNY, Baruch Coll, Paul H Chook Dept Informat Syst & Stat, One Baruch Way, New York, NY 10010 USA
关键词
Censored data; Empirical likelihood; Order-restricted inference; Stochastic ordering; LIKELIHOOD-BASED TESTS; EMPIRICAL LIKELIHOOD; NONPARAMETRIC-ESTIMATION; RATIO TEST; CONSTRAINTS;
D O I
10.1007/s10463-018-0694-5
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 paper, we develop an empirical likelihood-based test for the presence of stochastic ordering under censoring in the k-sample case. The proposed test statistic is formed by taking the supremum of localized empirical likelihood ratio test statistics. Its asymptotic null distribution has a simple representation in terms of a standard Brownian motion process. Through simulations, we show that it outperforms in terms of power existing methods for the same problem at all the distributions that we consider. A real-life example is used to illustrate the applicability of this new test.
引用
收藏
页码:451 / 470
页数:20
相关论文
共 50 条