A Robust Test in the Multivariate Two-Sample Location Problem

Ravinder Nath, Benjamin S. Duran

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

A frequently employed test for the equality of two mean vectors is Hotelling's two-sample T2, which assumes multinormality of the populations. When the multinormality assumption is violated, modified versions of T2 and/or nonparametric statistics are utilized. In this paper, it is shown that if Hotelling's T2 statistic is computed as a function of the ranks (instead of the original observations), the resulting statistic is a monotone function of a nonparametric statistic. Just as T2 reduces to the square of a t-statistic in univariate distributions, the rank T2 reduces to a rank transform statistic TR proposed by Conover and Iman (1981). For several bivariate distributions, Monte Carlo results are presented, which suggest robustness of rank T2.

Original languageEnglish
Pages (from-to)225-249
Number of pages25
JournalAmerican Journal of Mathematical and Management Sciences
Volume3
Issue number3
DOIs
StatePublished - 1983
Externally publishedYes

Fingerprint

Two-sample Problem
Robust Tests
Location Problem
Statistic
Statistics
Nonparametric Statistics
Hotelling's T2
Bivariate Distribution
Monotone Function
Univariate
Equality
Transform
Robustness
Location problem
Nonparametric statistics
Hotelling

All Science Journal Classification (ASJC) codes

  • Business, Management and Accounting(all)
  • Applied Mathematics

Cite this

A Robust Test in the Multivariate Two-Sample Location Problem. / Nath, Ravinder; Duran, Benjamin S.

In: American Journal of Mathematical and Management Sciences, Vol. 3, No. 3, 1983, p. 225-249.

Research output: Contribution to journalArticle

@article{44680d5cc8e94f55b54ad364c5c2c82d,
title = "A Robust Test in the Multivariate Two-Sample Location Problem",
abstract = "A frequently employed test for the equality of two mean vectors is Hotelling's two-sample T2, which assumes multinormality of the populations. When the multinormality assumption is violated, modified versions of T2 and/or nonparametric statistics are utilized. In this paper, it is shown that if Hotelling's T2 statistic is computed as a function of the ranks (instead of the original observations), the resulting statistic is a monotone function of a nonparametric statistic. Just as T2 reduces to the square of a t-statistic in univariate distributions, the rank T2 reduces to a rank transform statistic TR proposed by Conover and Iman (1981). For several bivariate distributions, Monte Carlo results are presented, which suggest robustness of rank T2.",
author = "Ravinder Nath and Duran, {Benjamin S.}",
year = "1983",
doi = "10.1080/01966324.1983.10737125",
language = "English",
volume = "3",
pages = "225--249",
journal = "American Journal of Mathematical and Management Sciences",
issn = "0196-6324",
publisher = "American Sciences Press Inc.",
number = "3",

}

TY - JOUR

T1 - A Robust Test in the Multivariate Two-Sample Location Problem

AU - Nath, Ravinder

AU - Duran, Benjamin S.

PY - 1983

Y1 - 1983

N2 - A frequently employed test for the equality of two mean vectors is Hotelling's two-sample T2, which assumes multinormality of the populations. When the multinormality assumption is violated, modified versions of T2 and/or nonparametric statistics are utilized. In this paper, it is shown that if Hotelling's T2 statistic is computed as a function of the ranks (instead of the original observations), the resulting statistic is a monotone function of a nonparametric statistic. Just as T2 reduces to the square of a t-statistic in univariate distributions, the rank T2 reduces to a rank transform statistic TR proposed by Conover and Iman (1981). For several bivariate distributions, Monte Carlo results are presented, which suggest robustness of rank T2.

AB - A frequently employed test for the equality of two mean vectors is Hotelling's two-sample T2, which assumes multinormality of the populations. When the multinormality assumption is violated, modified versions of T2 and/or nonparametric statistics are utilized. In this paper, it is shown that if Hotelling's T2 statistic is computed as a function of the ranks (instead of the original observations), the resulting statistic is a monotone function of a nonparametric statistic. Just as T2 reduces to the square of a t-statistic in univariate distributions, the rank T2 reduces to a rank transform statistic TR proposed by Conover and Iman (1981). For several bivariate distributions, Monte Carlo results are presented, which suggest robustness of rank T2.

UR - http://www.scopus.com/inward/record.url?scp=0040593598&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0040593598&partnerID=8YFLogxK

U2 - 10.1080/01966324.1983.10737125

DO - 10.1080/01966324.1983.10737125

M3 - Article

VL - 3

SP - 225

EP - 249

JO - American Journal of Mathematical and Management Sciences

JF - American Journal of Mathematical and Management Sciences

SN - 0196-6324

IS - 3

ER -