A multivariate version of Kendall's τ

被引：18

作者：

Choi, K

Marden, J ^{[1
]}

机构：

[1] Univ Illinois, Dept Stat, Champaign, IL 61820 USA

[2] HongiK Univ, Dept Math, Chochsion, South Korea

来源：

JOURNAL OF NONPARAMETRIC STATISTICS | 1998年 / 9卷 / 03期

基金：

美国国家科学基金会;

关键词：

Kendall's tau; multivariate signs; multiple regression; rank tests;

D O I：

10.1080/10485259808832746

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：

A multivariate version of Kendall's tau that is based on multivariate sign statistics is discussed. It has a nice visual interpretation and is easy to calculate. Tests of independence based on the statistic are shown to have goad efficiency and robustness properties relative to the usual least squares procedures.

引用

页码：261 / 293

页数：33

共 16 条

[1]

Abramowitz M., 1970, HDB MATH FUNCTIONS

[2]

BROWN BM, 1983, J ROY STAT SOC B MET, V45, P25

[3] MULTIVARIATE LOCATION ESTIMATION USING EXTENSION OF R-ESTIMATES THROUGH U-STATISTICS TYPE APPROACH [J].

CHAUDHURI, P .

ANNALS OF STATISTICS, 1992, 20 (02) :897-916

[4]

Chaudhuri P, 1996, J AM STAT ASSOC, V91, P862

[5] An approach to multivariate rank tests in multivariate analysis of variance [J].

Choi, K ;

Marden, J .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1997, 92 (440) :1581-1590

[6] NONPARAMETRIC ANALYSIS OF A GENERALIZED REGRESSION-MODEL - THE MAXIMUM RANK CORRELATION ESTIMATOR [J].

HAN, AK .

JOURNAL OF ECONOMETRICS, 1987, 35 (2-3) :303-316

[7]

Kendall MG, 1990, Correlation methods

[8]

M??tt??nen J., 1995, J NONPARAMETR STAT, V5, P201, DOI [10.1080/10485259508832643, DOI 10.1080/10485259508832643]

[9]

NARANJO JD, 1994, J ROY STAT SOC B MET, V56, P209

[10] A DISTRIBUTION-FREE MULTIVARIATE SIGN TEST BASED ON INTERDIRECTIONS [J].

RANDLES, RH .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1989, 84 (408) :1045-1050

← 1 2 →