Bandwidth selection for kernel conditional density estimation

被引：186

作者：

Bashtannyk, DM ^{[1
]}

Hyndman, RJ ^{[1
]}

机构：

[1] Monash Univ, Dept Econometr & Business Stat, Melbourne, Vic 3800, Australia

来源：

COMPUTATIONAL STATISTICS & DATA ANALYSIS | 2001年 / 36卷 / 03期

基金：

澳大利亚研究理事会;

关键词：

bandwidth selection; conditioning; density estimation; kernel smoothing;

D O I：

10.1016/S0167-9473(00)00046-3

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

We consider bandwidth selection for the kernel estimator of conditional density with one explanatory variable. Several bandwidth selection methods are derived ranging from fast rules-of-thumb which assume the underlying densities are known to relatively slow procedures which use the bootstrap. The methods are compared and a practical bandwidth selection strategy which combines the methods is proposed. The methods are compared using two simulation studies and a real data set. (C) 2001 Elsevier Science B.V. All rights reserved.

引用

页码：279 / 298

页数：20

共 14 条

[1] NEW LOOK AT STATISTICAL-MODEL IDENTIFICATION [J].

AKAIKE, H .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1974, AC19 (06) :716-723

[2]

Akaike H., 1973, 2 INT S INF THEOR, P268, DOI 10.1007/978-1-4612-1694-0_15

[3]

AZZALINI A, 1990, J R STAT SOC C-APPL, V39, P357

[4] Estimation of conditional densities and sensitivity measures in nonlinear dynamical systems [J].

Fan, JQ ;

Yao, QW ;

Tong, H .

BIOMETRIKA, 1996, 83 (01) :189-206

[5] Methods for estimating a conditional distribution function [J].

Hall, P ;

Wolff, RCL ;

Yao, QW .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1999, 94 (445) :154-163

[6]

Hardle W., 1991, Smoothing Techniques

[7]

Hyndman R. J., 1996, J COMPUT GRAPH STAT, V5, P315, DOI DOI 10.2307/1390887

[8]

HYNDMAN RJ, 1998, 1798 MON U DEP EC BU

[9]

*MATHS INC, 2000, S PLUS 2000 WIND

[10]

Rosenblatt M., 1969, MULTIVARIATE ANALYSI, P25

← 1 2 →