A Bayesian hierarchical model for longtitudinal data

The paper investigates a Bayesian hierarchical model for the analysis of longitudinal data from a randomized controlled clinical tuberculosis trial. Data for each subject are observed on thirteen time point of occasions of the trial. One of the features of the data set is that observations for some variables are missing for at least one time point. In the Bayesian approach, to estimate the model, we use the Gibbs sampler, which as well allows missing data for both the response and the explanatory variables to impute at each iteration of the algorithm, given some appropriate prior distributions