Williamson J M, Lipsitz S R, Kim K M
Centers for Disease Control and Prevention, Division of HIV/AIDS Prevention: Surveillance and Epidemiology, Atlanta, GA 30333, USA.
Comput Methods Programs Biomed. 1999 Jan;58(1):25-34. doi: 10.1016/s0169-2607(98)00063-7.
GEECAT and GEEGOR are two user-friendly SAS macros for the analysis of clustered, correlated categorical response data. Both programs implement methodology which extend the generalized estimating equation (GEE) approach of Liang and Zeger (Biometrika 73 (1986) 13-22). GEECAT and GEEGOR both use a first set of estimating equations to model the marginal response. With GEECAT, either correlated nominal or ordered categorical response data can be analyzed. The program GEEGOR employs a second set of estimating equations to model the association of ordered categorical responses within a cluster using the global odds ratio as a measure of association. The programs run on both mainframe computers and microcomputers. Examples are provided to illustrate the features of both programs.
GEECAT和GEEGOR是两个用户友好型的SAS宏程序，用于分析聚类的、相关的分类响应数据。这两个程序都实现了扩展Liang和Zeger（《生物统计学》73卷（1986年）第13 - 22页）广义估计方程（GEE）方法的方法。GEECAT和GEEGOR都使用第一组估计方程来对边际响应进行建模。使用GEECAT，可以分析相关的名义或有序分类响应数据。程序GEEGOR使用第二组估计方程，以全局优势比作为关联度量，对聚类内有序分类响应之间的关联进行建模。这些程序可在大型计算机和微型计算机上运行。文中提供了示例来说明这两个程序的特点。
