Estimation of conditional L1-median from dependent observations

被引：3

作者：

Berlinet, A ^{[1
]}

Cadre, B ^{[1
]}

Gannoun, A ^{[1
]}

机构：

[1] Univ Montpellier 2, Dept Math, F-34095 Montpellier 05, France

来源：

STATISTICS & PROBABILITY LETTERS | 2001年 / 55卷 / 04期

关键词：

conditional L-1-median; non-parametric estimation; non-i.i.d; observations;

D O I：

10.1016/S0167-7152(01)00046-3

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：

Let Q be a conditional distribution with L-1-median mu and let [Q(n)] be a sequence of conditional distributions converging in some sense to Q Then the L-1-medians {mu (n)} of distributions {Q(n)}are natural estimates of mu. In the case where {Q(n)} is a sequence of kernel estimates we give conditions ensuring that {mu (n)} is a well-defined sequence of continuous functions converging to It uniformly on compact sets. (C) 2001 Elsevier Science B.V. All rights reserved.

引用

页码：353 / 358

页数：6

共 12 条

[1]

[Anonymous], 1989, REAL ANAL PROBABILIT

[2]

BERLINET A, 1998, UNPUB NONPARAMETRIC

[3] Asymptotic distribution of data-driven smoothers in density and regression estimation under dependence [J].

Boente, G ;

Fraiman, R .

CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE, 1995, 23 (04) :383-397

[4]

Bosq D., 1987, THEORIE ESTIMATION F

[5] ALMOST COMPLETE CONVERGENCE PROPERTIES OF THE KERNEL PREDICTOR [J].

COLLOMB, G .

ZEITSCHRIFT FUR WAHRSCHEINLICHKEITSTHEORIE UND VERWANDTE GEBIETE, 1984, 66 (03) :441-460

[6]

DOUKHAN P, 1995, LECT NOTES STAT, V85

[7]

Gannoun A., 1990, PUBLICATIONS I STAT, VXXXV, P11

[8]

GYORFI L, 1989, LECT NOTES STAT, V60

[9]

Hardle W., 1990, APPL NONPARAMETRIC R, DOI DOI 10.1017/CCOL0521382483

[10]

HAREL M, 1995, AM J MATH MANAGEMENT, V15, P255

← 1 2 →