Comparison of hierarchical likelihood versus orthodox best linear unbiased predictor approaches for frailty models

被引:26
作者
Ha, ID [1 ]
Lee, Y
机构
[1] Daegu Haany Univ, Fac Informat Sci, Gyongsan 712715, South Korea
[2] Seoul Natl Univ, Dept Stat, Seoul 151742, South Korea
关键词
best linear unbiased predictor; frailty model; hierarchical likelihood; profile likelihood; random effect;
D O I
10.1093/biomet/92.3.717
中图分类号
Q [生物科学];
学科分类号
07 ; 0710 ; 09 ;
摘要
Hierarchical likelihood provides a statistically efficient procedure for frailty models. Recently, a method using the computationally attractive orthodox best linear unbiased predictor has been proposed; this uses Pearson-type estimation. We compare both approaches and discuss their relative merits. With semiparametric frailty models difficulties can arise for the orthodox method, if the number of nuisance parameters increases with the sample size. This difficulty is avoided by the use of the hierarchical-likelihood method.
引用
收藏
页码:717 / 723
页数:7
相关论文
共 8 条
[1]  
Breslow N. E., 1972, J. R. Stat. Soc. Ser. B. Stat. Methodol., V34, P216, DOI [10.1111/j.2517-6161.1972.tb00900.x, DOI 10.1111/J.2517-6161.1972.TB00900.X]
[2]   Hierarchical likelihood approach for frailty models [J].
Ha, ID ;
Lee, Y ;
Song, JK .
BIOMETRIKA, 2001, 88 (01) :233-243
[3]  
HA ID, 2003, J COMPUTATIONAL GRAP, V12, P663
[4]  
Lee Y, 1996, J ROY STAT SOC B MET, V58, P619
[5]   Hierarchical generalised linear models: A synthesis of generalised linear models, random-effect models and structured dispersions [J].
Lee, Y ;
Nelder, JA .
BIOMETRIKA, 2001, 88 (04) :987-1006
[6]   Random effects Cox models: A Poisson modelling approach [J].
Ma, RJ ;
Krewski, D ;
Burnett, RT .
BIOMETRIKA, 2003, 90 (01) :157-169
[7]   A comparison of some random effect models for parameter estimation in recurrent events [J].
Ng, ETM ;
Cook, RJ .
MATHEMATICAL AND COMPUTER MODELLING, 2000, 32 (1-2) :11-26
[8]   Estimation of multivariate frailty models using penalized partial likelihood [J].
Ripatti, S ;
Palmgren, J .
BIOMETRICS, 2000, 56 (04) :1016-1022