Improved approximations for multilevel models with binary responses

被引:319
作者
Goldstein, H
Rasbash, J
机构
[1] Institute of Education, London
[2] Institute of Education, University of London, London, WC1H 0AL
来源
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES A-STATISTICS IN SOCIETY | 1996年 / 159卷
关键词
binary response; generalized linear model; hierarchical data; marginal model; multilevel model; quasi-likelihood; unit-specific model;
D O I
10.2307/2983328
中图分类号
O1 [数学]; C [社会科学总论];
学科分类号
03 ; 0303 ; 0701 ; 070101 ;
摘要
This paper discusses the use of improved approximations for the estimation of generalized linear multilevel models where the response is a proportion. Simulation studies by Rodriguez and Goldman have shown that in extreme situations large biases can occur, most notably when the response is binary, the number of level 1 units per level 2 unit is small and the underlying random parameter values are large. An improved approximation is introduced which largely eliminates the biases in the situation described by Rodriguez and Goldman.
引用
收藏
页码:505 / 513
页数:9
相关论文
共 18 条
  • [1] AYIS SAM, 1995, THESIS U SOUTHAMPTON
  • [2] MARGINAL MAXIMUM-LIKELIHOOD ESTIMATION OF ITEM PARAMETERS - APPLICATION OF AN EM ALGORITHM
    BOCK, RD
    AITKIN, M
    [J]. PSYCHOMETRIKA, 1981, 46 (04) : 443 - 459
  • [3] BRESLOW NE, 1995, BIOMETRIKA, V82, P81
  • [4] APPROXIMATE INFERENCE IN GENERALIZED LINEAR MIXED MODELS
    BRESLOW, NE
    CLAYTON, DG
    [J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1993, 88 (421) : 9 - 25
  • [5] Diggle P., 2002, Analysis of longitudinal data
  • [6] Efron B, 1994, INTRO BOOTSTRAP, DOI DOI 10.1201/9780429246593
  • [7] GOLDSTEIN H, 1989, BIOMETRIKA, V76, P622
  • [8] GOLDSTEIN H, 1991, BIOMETRIKA, V78, P45
  • [9] GOLDSTEIN H, 1986, BIOMETRIKA, V73, P43
  • [10] Goldstein H., 2010, Multilevel statistical models, V4th