A SIMPLE ROOT N BANDWIDTH SELECTOR

被引：79

作者：

JONES, MC

MARRON, JS

PARK, BU

机构：

[1] SEOUL NATL UNIV,DEPT COMP SCI & STAT,SEOUL 151,SOUTH KOREA

[2] IBM CORP,DIV RES,ARMONK,NY 10504

[3] UNIV NORTH CAROLINA,DEPT STAT,CHAPEL HILL,NC 27599

来源：

ANNALS OF STATISTICS | 1991年 / 19卷 / 04期

关键词：

BANDWIDTH FACTORIZATION; BANDWIDTH SELECTION; DENSITY ESTIMATION; KERNEL ESTIMATORS; RATES OF CONVERGENCE; SMOOTHED CROSS-VALIDATION;

D O I：

10.1214/aos/1176348378

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：

The asymptotically best bandwidth selectors for a kernel density estimator currently require the use of either unappealing higher order kernel pilot estimators or related Fourier transform methods. The point of this paper is to present a methodology which allows the fastest possible rate of convergence with the use of only nonnegative kernel estimators at all stages of the selection process. The essential idea is derived through careful study of factorizations of the pilot bandwidth in terms of the original bandwidth.

引用

页码：1919 / 1932

页数：14

共 20 条

[1]

[Anonymous], 1988, EMPEC, DOI DOI 10.1007/BF01972448

[2]

CHUI ST, 1991, ANN STAT, V19, P1883

[3]

Eubank R.L., 1988, SPLINE SMOOTHING NON

[4] A FLEXIBLE AND FAST METHOD FOR AUTOMATIC SMOOTHING [J].

GASSER, T ;

KNEIP, A ;

KOHLER, W .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1991, 86 (415) :643-652

[5]

HALL P, 1991, J ROY STAT SOC B MET, V53, P245

[6] EXTENT TO WHICH LEAST-SQUARES CROSS-VALIDATION MINIMIZES INTEGRATED SQUARE ERROR IN NONPARAMETRIC DENSITY-ESTIMATION [J].

HALL, P ;

MARRON, JS .

PROBABILITY THEORY AND RELATED FIELDS, 1987, 74 (04) :567-581

[7] ESTIMATION OF INTEGRATED SQUARED DENSITY DERIVATIVES [J].

HALL, P ;

MARRON, JS .

STATISTICS & PROBABILITY LETTERS, 1987, 6 (02) :109-115

[8]

HALL P, 1991, BIOMETRIKA, V78, P263

[9]

HALL P, 1991, UNPUB SMOOTHED CROSS

[10]

HALL P, 1991, IN PRESS PROBAB THEO

← 1 2 →