Comparisons between largest order statistics from multiple-outlier models

被引:28
|
作者
Balakrishnan, Narayanaswamy [1 ]
Torrado, Nuria [2 ]
机构
[1] McMaster Univ, Dept Math & Stat, Hamilton, ON, Canada
[2] Univ Coimbra, Ctr Math, EC Santa Cruz, P-3001501 Coimbra, Portugal
基金
加拿大自然科学与工程研究理事会;
关键词
multiple-outlier models; stochastic orderings; proportional hazard rate models; heterogeneous random variables; order statistics; LIKELIHOOD RATIO; SAMPLES; SPACINGS;
D O I
10.1080/02331888.2015.1038268
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 work, we discuss stochastic comparisons between the largest order statistics from multiple-outlier models when the numbers of independent and identically distributed random variables are different. That is, if X-n:n(p,q) is the largest order statistic among X-1, ... , X-p, Xp+1, ... , X-n and X-n*:n* (p*,q*) is the largest order statistic among X-1, ... , X-p*, Xp*+1, ... , X-n*, where q = n - p and q* = n* - p*, we discuss whether X-n:n(p,q) and X-n*:n* (p*,q*) are ordered in some stochastic sense.
引用
收藏
页码:176 / 189
页数:14
相关论文
共 50 条