A SAS MACRO FOR STEPWISE CORRELATED BINARY REGRESSION

Several regression methods have been proposed for the analysis of corr elated binary data, but none deals with the selection of covariates wh en there exist a large number of potentially relevant covariates. We p resent a SAS macro based on a stepwise selection procedure for the ana lysis of correlated binary data. Using regression methods based on gen eralized estimating equations originally proposed by Liang and Zeger [ 1] and extended by Prentice [2], we describe a score test for forward selection, a Wald's test for backward elimination, and a test for mode l adequacy based on generalized scores, The methodology and the accomp anying computer macro program written in SAS IML are illustrated with data from a prospective study of functional decline in the activities of daily living in a group of elderly patients.