Bounded influence regression using high breakdown scatter matrices

被引：0

作者：

Croux, C

Van Aelst, S

Dehon, C

机构：

[1] Katholieke Univ Leuven, Dept Appl Econ, B-3000 Louvain, Belgium

[2] Univ Ghent, Dept Appl Math & Comp Sci, B-9000 Ghent, Belgium

[3] Univ Libre Brussels, ECARES, B-1050 Brussels, Belgium

来源：

ANNALS OF THE INSTITUTE OF STATISTICAL MATHEMATICS | 2003年 / 55卷 / 02期

关键词：

Fisher-consistency; influence function; robust regression; S-estimators; scatter matrices;

D O I：

暂无

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 estimate the parameters of a regression model using S-estimators of multivariate location and scatter. The approach is proven to be Fisher-consistent, and the influence functions are derived. The corresponding asymptotic variances are obtained and it is shown how they can be estimated in practice. A comparison with other recently proposed robust regression estimators is made.

引用

页码：265 / 285

页数：21

共 50 条

[41] Econometric applications of high-breakdown robust regression techniques
Zaman, A
Rousseeuw, PJ
Orhan, R
ECONOMICS LETTERS, 2001, 71 (01) : 1 - 8
[42] High-breakdown point estimation of some regression models
Sapra, SK
APPLIED ECONOMICS LETTERS, 2003, 10 (14) : 875 - 878
[43] The package: nonparametric regression using local rotation matrices in
Di Marzio, Marco
Fensore, Stefania
Lafratta, Giovanni
Taylor, Charles C.
JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION, 2021, 91 (11) : 2289 - 2306
[44] HIGH BREAKDOWN-POINT AND HIGH-EFFICIENCY ROBUST ESTIMATES FOR REGRESSION
YOHAI, VJ
ANNALS OF STATISTICS, 1987, 15 (02): : 642 - 656
[45] Fast algorithms for computing high breakdown covariance matrices with missing data
Copt, S
Victoria-Feser, MP
THEORY AND APPLICATION OF RECENT ROBUST METHODS, 2004, : 71 - 82
[46] Assessing Local Influence in Linear Regression Models with Specific Dispersion Matrices
张波
唐瑛
云南大学学报(自然科学版), 1998, (06)
[47] BOUNDED-INFLUENCE REGRESSION VIA LOCAL MINIMAX MEAN SQUARED ERROR
SAMAROV, AM
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1985, 80 (392) : 1032 - 1040
[48] Bounded Influence Support Vector Regression for Robust Single-Model Estimation
Dufrenois, Franck
Colliez, Johan
Hamad, Denis
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2009, 20 (11): : 1689 - 1706
[49] ROBUST LINEAR-REGRESSION VIA BOUNDED INFLUENCE M-ESTIMATORS
CHENG, CL
JOURNAL OF MULTIVARIATE ANALYSIS, 1992, 40 (01) : 158 - 171
[50] SOME SMALL-SAMPLE RESULTS ON A BOUNDED INFLUENCE RANK REGRESSION METHOD
WILCOX, RR
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 1995, 24 (04) : 881 - 888

← 1 2 3 4 5 →