Bayesian Modelling of Exponentiated Weibull Generated Family for Interval-censored Data with rstan

The purpose of this paper is to fit the Exponentiated Weibull generated (EW-G) family to interval-censored survival data in the Bayesian environment, which is an extended version of the Weibull-G family and generated by Exponentiated Weibull random variable. Based on the EW-G family, three special models, namely Exponentiated Weibull Exponential, Exponentiated Weibull Weibull, and Exponentiated Weibull Rayleigh are applied for the analysis. The models are fitted in R using the probabilistic programming language Stan, which offers full Bayesian inference through Hamiltonian Monte Carlo algorithm and their extension No-U-turn sampler algorithm. Stan codes for the analysis are provided. We compared the model with the leave-one-out-cross validation information criterion, and we found that simpler models with fewer parameters give a better fitting to the data.