Annals of Data Science
https://doi.org/10.1007/s40745-019-00228-1
Bayesian Survival Analysis of Type I General Exponential
Distributions
Mohammed H. AbuJarad
1
· Eman S. A. AbuJarad
2
· Athar Ali Khan
1
Received: 30 April 2019 / Revised: 22 June 2019 / Accepted: 13 July 2019
© Springer-Verlag GmbH Germany, part of Springer Nature 2019
Abstract
This article aims at generalizing two distribution by means of, exponentiated exponen-
tial and Weibull distribution. The researchers have employed three and four parameters
life model called the Type I General Exponential exponentiated exponential distribu-
tion and Type I General Exponential Weibull distribution. Survival and hazard rate
functions were provided for these two models. To fit these models into survival and
hazard rate functions, we adopted the Bayesian approach. For illustration, a real sur-
vival data set has been employed. Application is carried out by R and Stan. Finally,a
comparison between these two models is made by using loo package to find the best
model and simulation.
Keywords Posterior · Simulation · Loo · RStan · Bayesian Inference · WAIC · R ·
HMC
1 Introduction
Hamedani et al. [1] constructed new generate extended families. they present a new
class of distributions called the Type I General Exponential (TIGE) family of dis-
tributions. In this paper, aim is to fit the Type I General Exponential exponentiated
exponential distribution (TIGEEE) and Type I General Exponential Weibull distribu-
tion (TIGEW) using a Bayesian approach and this distribution has an important role
in lifetime modelling. Statistical methods for lifetimes data analysis have continued to
B Mohammed H. AbuJarad
m.jarad@gu.edu.ps
Eman S. A. AbuJarad
emanjarad2@gmail.com
Athar Ali Khan
atharkhan1962@gmail.com
1
Department of Statistics and Operations Research, AMU, Aligarh 202002, India
2
Department of Mathematics, AMU, Aligarh 202002, India
123