Estimation via EM-algorithm in a bivariate random effects model

This paper considers estimation in a bivariate:random-effects model allowing for arbitrary measurement times and variation in the number of observations for different individuals in the context of longitudinal studies. Two different structures for the covariance matrix of measurement error are considered, uncorrelated error between responses and correlation; of error terms at the same measurement times. The estimation of parameters for this model is via the EM-Algorithm. We derive the set of equations for both ML and REML estimation when the observed data consists of complete pairs. These equations are encoded in a SAS Macro utilizing SAS/IML for implementation crf the methodology. This is illustrated with an example from AIDS clinical trials.