Abstract

Competing risk situations occur when the systems/subjects under study are subjected to more than one cause of failure. In the present study, we propose Type-II hybrid censored competing risk analysis with masked system failure time data. It is assumed that the lifetime distribution of competing causes of failures follow Lindley distribution. We derive maximum likelihood and Bayes estimates of the model parameters. We also provide asymptotic and two bootstrap (boot-p and boot-t) confidence intervals of the model parameters. Markov Chain Monte Carlo technique such as Gibbs sampler is employed for Bayesian estimation. A simulation study is conducted to check the performances of the considered estimation methods. A masked competing risk real dataset is analyzed for illustrative purpose.

Author: Reetu Goel and Bhupendra Singh

Received on: March, 2021

Accepted on: November, 2021