EFFICIENT HIGH-BREAKDOWN M-ESTIMATORS OF SCALE

被引：20

作者：

CROUX, C ^{[1
]}

机构：

[1] UNIV ANTWERP,DEPT MATH & COMP SCI,B-2610 ANTWERP,BELGIUM

来源：

STATISTICS & PROBABILITY LETTERS | 1994年 / 19卷 / 05期

关键词：

BREAKDOWN POINT; INFLUENCE FUNCTION; ROBUSTNESS; SCALE ESTIMATION;

D O I：

10.1016/0167-7152(94)90005-1

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 the location model there exist M-estimators that combine a 50% breakdown point with an arbitrarily high efficiency. In this article we show that this is also the case for M-estimators of scale at normal models, although all of the well-known M-estimators of scale have a rather low efficiency there.

引用

页码：371 / 379

页数：9

共 7 条

[1]

HAMPEL FR, 1986, ROBUST STATISTICS AP

[2] LOWER BOUNDS FOR CONTAMINATION BIAS - GLOBALLY MINIMAX VERSUS LOCALLY LINEAR-ESTIMATION [J].

HE, XM ;

SIMPSON, DG .

ANNALS OF STATISTICS, 1993, 21 (01) :314-337

[3] ON THE OPTIMALITY OF S-ESTIMATORS [J].

HOSSJER, O .

STATISTICS & PROBABILITY LETTERS, 1992, 14 (05) :413-419

[4]

Huber PJ., 1981, ROBUST STATISTICS

[5]

MARTIN RD, 1993, ANN STAT, V22, P991

[6] ALTERNATIVES TO THE MEDIAN ABSOLUTE DEVIATION [J].

ROUSSEEUW, PJ ;

CROUX, C .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1993, 88 (424) :1273-1283

[7]

ROUSSEEUW PJ, 1984, LECTURE NOTES STATIS, V26

← 1 →