Multiple Imputation for Longitudinal Data Under a Bayesian Multilevel Model
Hakan Demirtaş · Communication in Statistics- Theory and Methods · 2009
In this article, I establish a connection between Bayesian random-coefficient pattern-mixture models that were described by Demirtas (Citation2005), and the idea of converting binary and ordinal longitudinal outcomes to multivariate normal outcomes in a sensible way so that re-conversion to the original scale yields the original specified marginal expectations and correlations after performing multiple imputation (Demirtas and Hedeker, Citation2007, Citation2008a). I also illustrate the use of these methods via a real data set from schizophrenia research.