A new signed rank test based on slopes of vectors for bivariate location problems

Sunil Mathur, Mohammad B. Sepehrifar

Research output: Contribution to journalArticle

Abstract

The proposed test is based on the slopes obtained by using two vectors. The test statistic does not depend on the covariance structure of the population, and is scale-invariant and robust to outliers. The asymptotic relative efficiency of the proposed test statistic with respect to Hotelling's T 2 indicates superior performance of the proposed test statistic under heavy-tailed distributions. A complete comparison of the proposed test with some of the existing test statistics is also provided. For non-normal distributions, a Monte Carlo simulation study shows that the proposed test statistic performs better than most of the existing test statistics compared here for almost all the shifts in the location. Application of the test is also illustrated using a real-life bivariate data set.

Original languageEnglish (US)
Pages (from-to)72-84
Number of pages13
JournalStatistical Methodology
Volume10
Issue number1
DOIs
StatePublished - Jan 1 2013

Fingerprint

Rank Test
Location Problem
Signed
Test Statistic
Slope
Non-normal Distribution
Asymptotic Relative Efficiency
Heavy-tailed Distribution
Scale Invariant
Covariance Structure
Outlier
Monte Carlo Simulation
Simulation Study

Keywords

  • Application
  • Bivariate test
  • Location asymptotic relative efficiency
  • Power

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

A new signed rank test based on slopes of vectors for bivariate location problems. / Mathur, Sunil; Sepehrifar, Mohammad B.

In: Statistical Methodology, Vol. 10, No. 1, 01.01.2013, p. 72-84.

Research output: Contribution to journalArticle

Mathur, Sunil ; Sepehrifar, Mohammad B. / A new signed rank test based on slopes of vectors for bivariate location problems. In: Statistical Methodology. 2013 ; Vol. 10, No. 1. pp. 72-84.
@article{6c16d40c5dd84a16896191e0da7c0e88,
title = "A new signed rank test based on slopes of vectors for bivariate location problems",
abstract = "The proposed test is based on the slopes obtained by using two vectors. The test statistic does not depend on the covariance structure of the population, and is scale-invariant and robust to outliers. The asymptotic relative efficiency of the proposed test statistic with respect to Hotelling's T 2 indicates superior performance of the proposed test statistic under heavy-tailed distributions. A complete comparison of the proposed test with some of the existing test statistics is also provided. For non-normal distributions, a Monte Carlo simulation study shows that the proposed test statistic performs better than most of the existing test statistics compared here for almost all the shifts in the location. Application of the test is also illustrated using a real-life bivariate data set.",
keywords = "Application, Bivariate test, Location asymptotic relative efficiency, Power",
author = "Sunil Mathur and Sepehrifar, {Mohammad B.}",
year = "2013",
month = "1",
day = "1",
doi = "10.1016/j.stamet.2012.07.001",
language = "English (US)",
volume = "10",
pages = "72--84",
journal = "Statistical Methodology",
issn = "1572-3127",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - A new signed rank test based on slopes of vectors for bivariate location problems

AU - Mathur, Sunil

AU - Sepehrifar, Mohammad B.

PY - 2013/1/1

Y1 - 2013/1/1

N2 - The proposed test is based on the slopes obtained by using two vectors. The test statistic does not depend on the covariance structure of the population, and is scale-invariant and robust to outliers. The asymptotic relative efficiency of the proposed test statistic with respect to Hotelling's T 2 indicates superior performance of the proposed test statistic under heavy-tailed distributions. A complete comparison of the proposed test with some of the existing test statistics is also provided. For non-normal distributions, a Monte Carlo simulation study shows that the proposed test statistic performs better than most of the existing test statistics compared here for almost all the shifts in the location. Application of the test is also illustrated using a real-life bivariate data set.

AB - The proposed test is based on the slopes obtained by using two vectors. The test statistic does not depend on the covariance structure of the population, and is scale-invariant and robust to outliers. The asymptotic relative efficiency of the proposed test statistic with respect to Hotelling's T 2 indicates superior performance of the proposed test statistic under heavy-tailed distributions. A complete comparison of the proposed test with some of the existing test statistics is also provided. For non-normal distributions, a Monte Carlo simulation study shows that the proposed test statistic performs better than most of the existing test statistics compared here for almost all the shifts in the location. Application of the test is also illustrated using a real-life bivariate data set.

KW - Application

KW - Bivariate test

KW - Location asymptotic relative efficiency

KW - Power

UR - http://www.scopus.com/inward/record.url?scp=84864392479&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864392479&partnerID=8YFLogxK

U2 - 10.1016/j.stamet.2012.07.001

DO - 10.1016/j.stamet.2012.07.001

M3 - Article

AN - SCOPUS:84864392479

VL - 10

SP - 72

EP - 84

JO - Statistical Methodology

JF - Statistical Methodology

SN - 1572-3127

IS - 1

ER -