期刊
BIOMETRIKA
卷 95, 期 3, 页码 773-778出版社
OXFORD UNIV PRESS
DOI: 10.1093/biomet/asn023
关键词
Akaike information criterion; conditional likelihood; longitudinal data; marginal likelihood; mixed-effects model; model selection
资金
- NIAID NIH HHS [R01 AI055290-03, R01 AI052765-01, R01 AI052765-04, N01 AI050020, R01 AI055290-02, R01 AI052765-03, R01 AI052765-02, R01 AI059773-02, R01 AI062247-03, R01 AI055290, R01 AI052765, R01 AI062247-02, R01 AI052765-05, R01 AI055290-04, R01 AI055290-01, R01 AI055290-05, R01 AI055290-06] Funding Source: Medline
The conventional model selection criterion, the Akaike information criterion, AIC, has been applied to choose candidate models in mixed-effects models by the consideration of marginal likelihood. Vaida & Blanchard (2005) demonstrated that such a marginal AIC and its small sample correction are inappropriate when the research focus is on clusters. Correspondingly, these authors suggested the use of conditional AIC. Their conditional AIC is derived under the assumption that the variance-covariance matrix or scaled variance-covariance matrix of random effects is known. This note provides a general conditional AIC but without these strong assumptions. Simulation studies show that the proposed method is promising.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据