Abstract
The assumption of independence of causes for modeling a competing risks scenario is not presumable always. In this article, cause dependent competing risks model has been analyzed under Marshall-Olkin set-up. A Marshall-Olkin generalized lifetime distribution has been established to address a competing risks model. The essential statistical properties have been derived for this distribution to utilize in a competing risks model. The estimators of unknown simple/transformed parameters are obtained by using two approaches: maximum likelihood method and Bayesian estimation through reference prior. To examine the performance of the five models under generalized lifetime family, a simulation study has been performed. For illustration, two real data sets namely, prostate cancer and diabetic retinopathy study are considered. For both data sets, analysis is performed using the most suitable model from Marshall-Olkin generalized lifetime family.
Notes
1 MOBE: Marshall-Olkin Bivariate Exponential; MOBR: Marshall-Olkin Bivariate Rayleigh; MOBP: Marshall-Olkin Bivariate Pareto; MOBB: Marshall-Olkin Bivariate Burr; MOBW: Marshall-Olkin Bivariate Weibull.