Kernel weighted influence measures

被引：9

作者：

Hens, N

Aerts, M

Molenberghs, G

Thijs, H

Verbeke, G

机构：

[1] Limburgs Univ Ctr, Ctr Stat, B-3590 Diepenbeek, Belgium

[2] Katholieke Univ Leuven, Ctr Biostat, B-3000 Louvain, Belgium

来源：

COMPUTATIONAL STATISTICS & DATA ANALYSIS | 2005年 / 48卷 / 03期

关键词：

local influence; global influence; kernel weights; missing data; sensitivity analysis; weighted; likelihood;

D O I：

10.1016/j.csda.2004.02.010

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

To asses the sensitivity of conclusions to model choices in the context of selection models for non-random dropout, several methods have been developed. None of them are without limitations. A new method called kernel weighted influence is proposed. While global and local influence approaches look upon the influence of cases, this new method looks at the influence of types of observations. The basic idea is to combine the existing influence approaches with a non-parametric weighting scheme. The kernel weighted global influence offers a possible solution to the problem of masking, while the kernel weighted local influence can be seen as a tool to better understand the source of influence. (C) 2004 Elsevier B.V. All rights reserved.

引用

页码：467 / 487

页数：21

共 16 条

[1]

Cook R. D., 1982, Residuals and influence in regression

[2]

COOK RD, 1986, J ROY STAT SOC B MET, V48, P133

[3]

DIGGLE P, 1994, J ROY STAT SOC C, V43, P49

[4]

Kenward MG, 1998, STAT MED, V17, P2723, DOI 10.1002/(SICI)1097-0258(19981215)17:23&lt

[5]

2723::AID-SIM38&gt

[6]

3.0.CO

[7]

2-5

[8] Local influence in linear mixed models [J].

Lesaffre, E ;

Verbeke, G .

BIOMETRICS, 1998, 54 (02) :570-582

[9]

Little R.J., 1987, Statistical Analysis With Missing Data

[10] Influence analysis to assess sensitivity of the dropout process [J].

Molenberghs, G ;

Verbeke, C ;

Thijs, H ;

Lesaffre, E ;

Kenward, MG .

COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2001, 37 (01) :93-113

← 1 2 →