bayesian modeling of the dependence in longitudinal data via.pdf

Bayesian modeling of the dependence in longitudinal data via partial autocorrelations and marginal variances Y. Wang M. J. Daniels yanpin@ufl.edu mdaniels@stat.ufl.edu Department of Biostatistics Department of Statistics University of Florida University of Florida Summary Many parameters and positive-definiteness are two major obstacles in estimating and modelling a correlation matrix for longitudinal data. In addition, when longitudinal data is incomplete, incorrectly modelling the correlation matrix often results in bias in estimating mean regression parameters. In this paper, we introduce a flexible and parsimonious class of regression models for a covariance matrix parameterized using marginal variances and partial autocorrelations. The partial autocorrelations can freely vary in the interval (−1, 1) while maintaining positive defi- niteness of the correlation matrix so the regression parameters in these models will have no con- straints. We propose a class of priors for the regression coefficients and examine the importance of correctly modeling the correlation structure on estimation of longitudinal (mean) trajectories and the performance of the DIC is choosing the correct correlation model via simulations. The regression approach is illustrated on data from a longitudinal clinical trial. Some key words : Markov Chain Monte Carlo; Generalized linear model; Uniform prior and triangular prior. 1 1 Introduction Longitudinal data, measurements on the same subject over time, arise in many areas, from clin- ical trials to environmental studies. In such studies, to draw valid inference, the covariance between repeated observations on the same individual