On the weighted multivariate Wilcoxon rank regression estimate

Weihua Zhou, Jin Wang

Research output: Contribution to journalArticle

Abstract

Zhou (2010) introduced a multivariate Wilcoxon regression estimate which possesses some nice properties: computational ease, asymptotic normality and high efficiency. However, it is sensitive to the leverage points. To circumvent this problem, we propose a weighted multivariate Wilcoxon regression estimate. Under some regularity conditions, the asymptotic normality is established. We further study the robustness of the proposed estimate through the influence function. By properly choosing the weight functions, our results show that the corresponding estimate can have bounded influence function on both response and covariates.

Original languageEnglish (US)
Pages (from-to)704-713
Number of pages10
JournalStatistics and Probability Letters
Volume81
Issue number6
DOIs
StatePublished - Jun 2011

Fingerprint

Rank Regression
Regression Estimate
Influence Function
Asymptotic Normality
Leverage Points
Bounded Influence
Regularity Conditions
Weight Function
Estimate
High Efficiency
Covariates
Robustness

Keywords

  • Influence function
  • Multivariate regression
  • Primary
  • Rank estimate
  • Secondary
  • Wilcoxon

ASJC Scopus subject areas

  • Statistics, Probability and Uncertainty
  • Statistics and Probability

Cite this

On the weighted multivariate Wilcoxon rank regression estimate. / Zhou, Weihua; Wang, Jin.

In: Statistics and Probability Letters, Vol. 81, No. 6, 06.2011, p. 704-713.

Research output: Contribution to journalArticle

@article{c805f0268b0547a5bce9580089943ed8,
title = "On the weighted multivariate Wilcoxon rank regression estimate",
abstract = "Zhou (2010) introduced a multivariate Wilcoxon regression estimate which possesses some nice properties: computational ease, asymptotic normality and high efficiency. However, it is sensitive to the leverage points. To circumvent this problem, we propose a weighted multivariate Wilcoxon regression estimate. Under some regularity conditions, the asymptotic normality is established. We further study the robustness of the proposed estimate through the influence function. By properly choosing the weight functions, our results show that the corresponding estimate can have bounded influence function on both response and covariates.",
keywords = "Influence function, Multivariate regression, Primary, Rank estimate, Secondary, Wilcoxon",
author = "Weihua Zhou and Jin Wang",
year = "2011",
month = "6",
doi = "10.1016/j.spl.2011.02.022",
language = "English (US)",
volume = "81",
pages = "704--713",
journal = "Statistics and Probability Letters",
issn = "0167-7152",
publisher = "Elsevier",
number = "6",

}

TY - JOUR

T1 - On the weighted multivariate Wilcoxon rank regression estimate

AU - Zhou, Weihua

AU - Wang, Jin

PY - 2011/6

Y1 - 2011/6

N2 - Zhou (2010) introduced a multivariate Wilcoxon regression estimate which possesses some nice properties: computational ease, asymptotic normality and high efficiency. However, it is sensitive to the leverage points. To circumvent this problem, we propose a weighted multivariate Wilcoxon regression estimate. Under some regularity conditions, the asymptotic normality is established. We further study the robustness of the proposed estimate through the influence function. By properly choosing the weight functions, our results show that the corresponding estimate can have bounded influence function on both response and covariates.

AB - Zhou (2010) introduced a multivariate Wilcoxon regression estimate which possesses some nice properties: computational ease, asymptotic normality and high efficiency. However, it is sensitive to the leverage points. To circumvent this problem, we propose a weighted multivariate Wilcoxon regression estimate. Under some regularity conditions, the asymptotic normality is established. We further study the robustness of the proposed estimate through the influence function. By properly choosing the weight functions, our results show that the corresponding estimate can have bounded influence function on both response and covariates.

KW - Influence function

KW - Multivariate regression

KW - Primary

KW - Rank estimate

KW - Secondary

KW - Wilcoxon

UR - http://www.scopus.com/inward/record.url?scp=79952543812&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952543812&partnerID=8YFLogxK

U2 - 10.1016/j.spl.2011.02.022

DO - 10.1016/j.spl.2011.02.022

M3 - Article

AN - SCOPUS:79952543812

VL - 81

SP - 704

EP - 713

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

SN - 0167-7152

IS - 6

ER -