Bayesian variable selection and estimation in joint confirmatory factor analysis-Cox model

In this article, we propose the joint confirmatory factor analysis-Cox model to assess the effects of observed and latent risk factors on survival time. The Bayesian adaptive Lasso procedure is developed to simultaneously conduct estimation and variable selection for the proposed model. Nice features including the empirical performance of the proposed method are demonstrated by simulation studies. The proposed method is applied to analyze the bladder cancer data set obtained from the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute.