Selection of working correlation structure and best model in GEE analyses of longitudinal data
Selection of working correlation structure and best model in GEE analyses of longitudinal data
复制标题
DOI：
10.1080/03610910701539617
复制
发表时间：
2007-01-01
影响因子：
0.9
通讯作者：
Qian, Guoqi
中科院分区：
文献类型：
--
作者：
Cui, James;Qian, Guoqi
广义估计方程（GEE）方法是流行病学研究中纵向数据分析最常用的统计方法之一。该方法需要为受试者的结果变量的重复测量指定一个有效的相关结构。然而，由于GEE方法是在准似然理论的基础上发展起来的，因此，在GEE中选择最佳相关结构和最佳解释变量子集的统计标准是最近才出现的。基于极大似然的模型选择方法，如广泛使用的赤池信息准则（Akaike Information Criterion， AIC），并不直接适用于GEE。Pan（2001）提出了一种QIC选择方法，可以选择最佳的相关结构和解释变量的最佳子集。基于QIC方法，我们开发了一个计算程序来计算一系列不同分布、链接函数和相关结构的QIC值。这个程序是用Stata软件编写的。在本文中，我们介绍了这个程序，并通过几个有代表性的例子说明如何使用它来选择最简洁的模型在纵向数据的GEE分析中。
The Generalized Estimating Equations ( GEE) method is one of the most commonly used statistical methods for the analysis of longitudinal data in epidemiological studies. A working correlation structure for the repeated measures of the outcome variable of a subject needs to be specified by this method. However, statistical criteria for selecting the best correlation structure and the best subset of explanatory variables in GEE are only available recently because the GEE method is developed on the basis of quasi-likelihood theory. Maximum likelihood based model selection methods, such as the widely used Akaike Information Criterion (AIC), are not applicable to GEE directly. Pan (2001) proposed a selection method called QIC which can be used to select the best correlation structure and the best subset of explanatory variables. Based on the QIC method, we developed a computing program to calculate the QIC value for a range of different distributions, link functions and correlation structures. This program was written in Stata software. In this article, we introduce this program and demonstrate how to use it to select the most parsimonious model in GEE analyses of longitudinal data through several representative examples.