On power and necessary sample sizes of the Wilcoxon-Mann-Whitney test

被引：3

作者：

Frick, H ^{[1
]}

Rahlfs, VW ^{[1
]}

机构：

[1] Inst Datenanal & Versuchsplanung Gauting, D-82131 Gauting, Germany

来源：

COMMUNICATIONS IN STATISTICS-THEORY AND METHODS | 1998年 / 27卷 / 10期

关键词：

Mann-Whitney U-statistic; Birnbaum-Klose and Ury-Wiggins variance bounds; discrete null-variance;

D O I：

10.1080/03610929808832236

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：

The Wilcoxon-Mann-Whitney test is a frequently applied nonparametric procedure for testing the equality of two distributions. In this paper formulae for the necessary sample sizes and lower bounds for the power are derived applicable to arbitrary distributions. In particular, the case of discrete random variables is investigated. It is shown how the power bounds increase and the necessary sample sizes decrease when the number of possible outcomes decreases.

引用

页码：2445 / 2460

页数：16

共 7 条

[1] Nonparametric hypotheses and rank statistics for unbalanced factorial designs [J].

Akritas, MG ;

Arnold, SF ;

Brunner, E .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1997, 92 (437) :258-265

[2] BOUNDS FOR THE VARIANCE OF THE MANN-WHITNEY STATISTIC [J].

BIRNBAUM, ZW ;

KLOSE, OM .

ANNALS OF MATHEMATICAL STATISTICS, 1957, 28 (04) :933-945

[3] NONPARAMETRIC METHODS FOR STRATIFIED 2-SAMPLE DESIGNS WITH APPLICATION TO MULTICLINIC TRIALS [J].

BRUNNER, E ;

PURI, ML ;

SUN, S .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1995, 90 (431) :1004-1014

[4] 2 - SAMPLE RANK-TESTS IN GENERAL-MODELS [J].

BRUNNER, E ;

NEUMANN, N .

BIOMETRICAL JOURNAL, 1986, 28 (04) :395-402

[5]

HILGERS R, 1981, BIOMETR J, V23, P653

[6]

Lehmann EL, 1975, NONPARAMETRICS

[7] GENERAL UPPER BOUND ON VARIANCE OF WILCOXON-MANN-WHITNEY U-STATISTIC FOR SYMMETRIC DISTRIBUTIONS WITH SHIFT ALTERNATIVES [J].

URY, HK ;

WIGGINS, AD .

BRITISH JOURNAL OF MATHEMATICAL & STATISTICAL PSYCHOLOGY, 1976, 29 (NOV) :263-267

← 1 →