|
Abstract:
|
In many applications, it is of interest to assess the dependence structure in multivariate longitudinal data. For example, in studies of oxidative stress and reproductive hormones, high levels of oxidative stress in critical windows of the menstrual cycle may relate to altered hormone profiles. Discovering such dependence is challenging due to the dimensionality involved. Concatenating the random effects from component models for each response, dependence within and across longitudinal responses is characterized through a large random effects covariance matrix. Facing problems in estimating this matrix and conducting inferences on off-diagonal elements, we propose a Bayesian approach that relies on shrinkage priors for parameters in a modified Cholesky decomposition. An efficient Gibbs sampler is developed for posterior computation.
|
