Bayesian adaptive lasso for additive hazard regression with current status data.

Area of Science:

• Biostatistics
• Survival Analysis
• Statistical Modeling

Background:

• Variable selection is critical in survival analysis, but existing methods often overlook current status data.
• Most current variable selection procedures for survival models are based on the frequentist framework.

Purpose of the Study:

• To propose a novel Bayesian adaptive least absolute shrinkage and selection operator (LASSO) procedure for simultaneous variable selection and parameter estimation.
• To address the limitations of existing methods by focusing on current status data within an additive hazards model.

Main Methods:

• Developed a Bayesian adaptive LASSO procedure for variable selection and parameter estimation.
• Utilized efficient Markov chain Monte Carlo (MCMC) methods for posterior sampling and inference.
• Applied the method to current status data within an additive hazards model framework.

Main Results:

• The proposed Bayesian adaptive LASSO method demonstrated effective simultaneous variable selection and parameter estimation.
• Simulation studies confirmed the empirical performance of the developed procedure.
• The method was successfully applied to identify risk factors for heart failure in type 2 diabetes patients.

Conclusions:

• The proposed Bayesian adaptive LASSO offers a robust approach for variable selection in survival analysis with current status data.
• This method provides a valuable tool for identifying key risk factors in complex health studies.
• The study highlights the utility of Bayesian methods in addressing limitations of frequentist approaches in survival data analysis.