Using SAS/GENMOD Procedure to Fit GEE Regression Models

Researchers are often interested in analyzing data that arise from longitudinal studies. And estimating equations for generalized linear modeling of longitudinal data have attracted a great deal of attention over the last two decades. Liang and Zeger (1986) presented an approach, generalized estimating equations (GEEs) which extended from generalized linear models (GLMs) to a regression setting with correlated observations within subjects, to these problems. This paper provides briefly review the GLM and GEE methodologies, and illustrate its implementation with a home-care example using the GENMOD procedure in SAS/STAT software to solve GEE in the analysis of correlated data.