On necessary conditions for the weak consistency of minimum L-1-norm estimates in linear models

被引：3

作者：

Bai, ZD

Wu, Y

机构：

[1] NATL SUN YAT SEN UNIV,DEPT MATH APPL,KAOHSIUNG 80424,TAIWAN

[2] YORK UNIV,DEPT MATH & STAT,N YORK,ON M3J 1P3,CANADA

来源：

STATISTICS & PROBABILITY LETTERS | 1997年 / 34卷 / 02期

关键词：

minimum L-1 norm estimate; weak consistency; linear regression model; necessary condition;

D O I：

10.1016/S0167-7152(96)00182-4

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：

Consider the model y(i) = x'(i) beta(0) + e(i), i = 1,..., n. Under very weak conditions on the error distributions, it is shown that inf(\mu\=1) Sigma(i=1)(infinity) \mu'x(i)\ = infinity is a necessary condition for the weak consistency of a minimum L-1-norm estimate of beta(0), which cannot be further improved.

引用

页码：193 / 199

页数：7

共 16 条

[1] 2 STAGE LEAST ABSOLUTE DEVIATIONS ESTIMATORS
AMEMIYA, T
[J]. ECONOMETRICA, 1982, 50 (03) : 689 - 711
[2] ARCONES MA, 1996, 2 ORDER REPRESENTATI
[3] BAI Z, 1991, PROBABILITY STAT DES, P33
[4] BASSETT G, 1978, J AM STAT ASSOC, V57, P33
[5] Bloomfield P., 1983, LEAST ABSOLUTE DEVIA
[6] CEN X, 1993, STAT SINICA, V3, P9
[7] CHEN X, 1992, RES NOTES MATH SERIE, V258, P249
[8] CHEN X, 1992, PROBABILTIY STAT, P32
[9] STRONG CONSISTENCY OF M-ESTIMATES IN LINEAR-MODELS
CHEN, XR
WU, YH
[J]. JOURNAL OF MULTIVARIATE ANALYSIS, 1988, 27 (01) : 116 - 130
[10] ON A NECESSARY CONDITION FOR THE CONSISTENCY OF THE L1 ESTIMATES IN LINEAR-REGRESSION MODELS
CHEN, XR
WU, Y
[J]. COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 1993, 22 (03) : 631 - 639

← 1 2 →