Fitting Longitudinal Data with Hierarchical Generalized Linear Models

摘要：

We propose a class of hierarchical generalized linear models (HGLMs) with random dispersions in this paper, and focus on the properties of the L-N estimators for the fixed effect β in the extended Poisson-Gamma models which are typical hierarchical generalized linear models. Under the proper assumptions on response variables and some smoothing conditions, we obtain the strong consistency and the convergence rate of the L-N estimator based on the combination of L-N and quasi-likelihood. Furthermore, we also obtain its asymptotical normality based on the combination of the central limit theorem and the law of large numbers. At last, an example is presented for the illustration of the proposed model and the inference method.

关键词： H-likelihood L-N estimators Poisson-Gamma models strong consistency random effect

作者: Na Xia Zhong-Zhan Zhang

作者单位: College of Applied Sciences, Beijing University of Technology Beijing, 100022, P.R.China

会议类型: 国际会议

会议名称: 2006 International Conference on Design of Experiments and Its Applications(2006实验设计及其应用国际会议)

会议地点: 天津

会议语种:英文

在线出版日期: 2006-07-09（万方平台首次上网日期，不代表论文的发表时间）