Re-sampling methods for testing for location against unrestricted and ordered alternatives

被引：0
|
作者：
Babu, Gutti Jogesh [1 ]
Padmanabhan, A. R.
机构：
[1] Penn State Univ, Dept Stat, University Pk, PA 16802 USA
[2] Monash Univ, Dept Math & Stat, Clayton, Vic 3800, Australia
来源：
JOURNAL OF STATISTICAL PLANNING AND INFERENCE | 2007年 / 137卷 / 11期
关键词：
bootstrap; Mann-Whitney statistic; multi-sample F-K; med statistic;
D O I：
10.1016/j.jspi.2007.03.010
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：
Robust tests are proposed for unrestricted and ordered alternatives in the multi-sample problem without assuming homogeneity of scales and/or symmetry of the underlying distributions. The methodology consists of bootstrapping appropriately centered Mann-Whitney statistics. Data sets from Physics and Psychology illustrate the methodology. (C) 2007 Elsevier B.V. All rights reserved.
引用
收藏
页码：3261 / 3267
页数：7
相关论文
共 41 条
12345