Robustness of deepest regression

被引：29

作者：

Van Aelst, S ^{[1
]}

Rousseeuw, PJ ^{[1
]}

机构：

[1] Univ Instelling Antwerp, Dept Math & Comp Sci, B-2610 Wilrijk, Belgium

来源：

JOURNAL OF MULTIVARIATE ANALYSIS | 2000年 / 73卷 / 01期

关键词：

breakdown value; influence function; regression depth;

D O I：

10.1006/jmva.1999.1870

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：

In this paper we investigate the robustness properties of the deepest regression, a method for linear regression introduced by Rousseeuw and Hubert [6]. We show that the deepest regression functional is Fisher-consistent fur the conditional median, and has a breakdown value of 1/3 in all dimensions. We also derive its influence function, and compare it with sensitivity functions. (C) 2000 Academic Press.

引用

页码：82 / 106

页数：25

共 9 条

[1]

BAI Z, IN PRESS ANN STAT

[2]

HAMPEL FR, 1986, ROBOT STAT APPROACH

[3]

HE X, 1998, APPL STAT SCI, V3

[4]

Rousseeuw P. J., 1984, LECTURE NOTES STATIS, V26, P256

[5]

Rousseeuw PJ, 1999, J AM STAT ASSOC, V94, P388, DOI 10.2307/2670155

[6] Regression depth - Rejoinder [J].

Rousseeuw, PJ ;

Van Aelst, S ;

Hubert, M .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1999, 94 (446) :419-433

[7]

ROUSSEEUW PJ, 1995, COMPUTATION STAT, V10, P71

[8] LEAST MEDIAN OF SQUARES REGRESSION [J].

ROUSSEEUW, PJ .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1984, 79 (388) :871-880

[9]

TUKEY JW, 1975, 1974 P INT C MATH VA, V2, P523

← 1 →