A COMPUTER-PROGRAM FOR REGRESSION-ANALYSIS OF REPEATED-MEASURES USING GENERALIZED ESTIMATING EQUATIONS

被引:54
作者
DAVIS, CS
机构
[1] Department of Preventive Medicine and Environmental Health, University of Iowa, Iowa City
关键词
CORRELATED RESPONSES; DISCRETE DATA; GENERALIZED LINEAR MODELS; LONGITUDINAL DATA; QUASI-LIKELIHOOD;
D O I
10.1016/0169-2607(93)90046-N
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
RMGEE is an easy-to-use FORTRAN program for the analysis of repeated binary, count, and normally-distributed response variables using the generalized estimating equations approach of Liang and Zeger [1]. The program can be used when measurements are obtained at multiple time points from each subject or experimental unit, and also when the basic sampling unit is a group or cluster of subjects, and the response variable of interest is obtained from each subject within the cluster. It is not necessary for the number of repeated measurements to be the same for every experimental unit and missing data are easily accommodated. Both time-independent (cluster-specific) and time-dependent (occasion- or subject-specific) covariates are permitted. The program can be run on microcomputers, workstations, and mainframe computers. Three examples illustrating the usage and features of RMGEE are provided.
引用
收藏
页码:15 / 31
页数:17
相关论文
共 13 条