A DISCRETE SURVIVAL MODEL WITH RANDOM EFFECTS - AN APPLICATION TO TIME TO PREGNANCY

Time to pregnancy, the number of menstrual cycles it takes a couple to conceive, and various covariates have been collected among couples ul timately achieving conception. To assess the influence of the covariat es, we constructed a discrete survival model that allows time-dependen t covariates. A random effect was included to account for unobserved h eterogeneity. The collected waiting times are obtained through retrosp ective ascertainment and are analyzed as truncated data. Maximum likel ihood estimation was implemented by Fisher scoring through iteratively reweighted least squares.