Parametric proportional hazard models using the Bayesian approach with applications to healthcare data

The aim of this study is on using Bayesian inference to analyze right-censored healthcare data using Frechet and exponential baseline proportional hazard (PH) models. For the baseline hazard parameters, a gamma prior was used, and for the regression coefficients, normal priors were used. The exact form of the joint posterior distribution was obtained. Bayes estimators of the parameters are obtained using the Markov chain Monte Carlo (MCMC) simulation technique. Two real-survival data applications were analyzed by the Frechet PH model and the exponential PH model. The convergence diagnostic tests are presented. We found that the Frechet PH model was better than the exponential PH model because it is flexible and could be beneficial in analyzing survival data.