Bayesian Joint Modeling Analysis of Longitudinal Proportional and Survival Data

This paper focuses on a joint model to analyze longitudinal proportional and survival data. We utilize logit transformation the data employ partially linear mixed-effect model. With this model, we estimate unknown function of time using B-splines technique. Additionally, introduce centered Dirichlet process mixture (CDPMM) capture random effects, allowing for flexible distribution. The are assumed Cox hazard sharing effects is developed two types develop Bayesian Lasso (BLasso) approach that combines Gibbs sampler Metropolis–Hastings algorithm. proposed method allows estimation parameters selection significant covariates simultaneously. evaluate performance our methods through simulation studies also provide an illustration methodologies example from MA.5 research experiment.