Case deletion diagnostics in multilevel models

被引:25
作者
Shi, Lei [2 ]
Chen, Gemai [1 ,3 ]
机构
[1] Univ Calgary, Dept Math & Stat, Calgary, AB T2N 1N4, Canada
[2] Yunnan Univ Finance & Econ, Coll Stat & Math, Kunming 650221, Yunnan, Peoples R China
[3] SW Jiaotong Univ, Dept Appl Math, Chengdu, Sichuan, Peoples R China
基金
加拿大自然科学与工程研究理事会; 中国国家自然科学基金;
关键词
Multilevel models; Case deletion; Influential observations; Cook's distance;
D O I
10.1016/j.jmva.2008.01.023
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 [统计学]; 070103 [概率论与数理统计]; 0714 [统计学];
摘要
This paper studies case deletion diagnostics for multilevel models. Using subset deletion, diagnostic measures for identifying influential units at any level are developed for both fixed and random parameters. Two approximate update formulae are derived. The first formula uses one-step approximation, while the second formula also includes the impact of estimating the random parameter. Two examples are used to illustrate the methodology developed. (C) 2008 Elsevier Inc. All rights reserved.
引用
收藏
页码:1860 / 1877
页数:18
相关论文
共 17 条
[1]
Atkinson A.C., 2000, SPR S STAT, DOI 10.1007/978-1-4612-1160-0
[2]
Atkinson AC, 1998, J ROY STAT SOC B, V60, P521
[3]
Influence diagnostics for linear longitudinal models [J].
Banerjee, M ;
Frees, EW .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1997, 92 (439) :999-1005
[4]
Belsley D. A., 1980, REGRESSION DIAGNOSTI
[5]
Chatterjee S., 1988, Sensitivity Analysis in Linear Regression, DOI 10.1002/9780470316764
[6]
CASE-DELETION DIAGNOSTICS FOR MIXED MODELS [J].
CHRISTENSEN, R ;
PEARSON, LM ;
JOHNSON, W .
TECHNOMETRICS, 1992, 34 (01) :38-45
[7]
Cook D.R., 1999, APPL REGRESSION INCL
[8]
Cook R. D., 1982, Residuals and influence in regression
[9]
GOLDSTEIN H, 1989, BIOMETRIKA, V76, P622
[10]
GOLDSTEIN H, 1986, BIOMETRIKA, V73, P43