Statistical inference for competing risks model with Type-II generalized hybrid censored of inverse Lomax distribution with applications

Abstract

This article considered statistical inference based on independent competing risks data from inverse Lomax (ILo) distribution using Type-II generalized hybrid censored dataset. Time-to-failure under many causes of failure, which might be dependent or independent, defines the competing risks model. The situation of independent observed reasons of failure competing risks model is the main topic of this work. With Type-II generalized hybrid censored competing risks, the maximum likelihood estimates (MLE) of the unknown model parameters and the life parameters (reliability and hazard rate functions) are derived using the Newton–Raphson method, and the bootstrap approach is studied. By utilizing the MLEs’ asymptotic normality property and the observed Fisher information matrix, approximate confidence intervals are created. Bayes estimates and highest posterior density (HPD) credible interval also are calculated under gamma prior distribution with the application of the Markov chain Monte Carlo (MCMC) approach yields both loss functions that are symmetric and asymmetric. Furthermore, a Monte Carlo simulation is run to evaluate the efficacy of the proposed methods. A set of real-world data is examined to illustrate the feasibility and suitability of the proposed methods.

Keywords

• Competing risks model
• Inverse Lomax distribution
• Bayesian inference
• The symmetric and asymmetric loss functions
• Metropolis–Hastings algorithm
• Maximum likelihood estimation