Statistical inference for dependent competing risks data under improve adaptive Type-II progressive hybrid censored partially accelerate life test

This paper presents a statistical analysis of dependent competing risks data modeled using the Marshall-Olkin bivariate Weibull distribution, under an improved adaptive Type-II progressive hybrid censored partially accelerated life test (IAT-II PHCS). We derive the maximum likelihood estimations (MLEs) for the unknown parameters and rigorously prove their existence and uniqueness. Asymptotic confidence intervals (ACIs) for the parameters are constructed using asymptotic theory and the delta method. In addition, Bayesian estimates are obtained using a gamma-Dirichlet prior distribution, with highest posterior density credible intervals (HPD CIs) computed through the importance sampling method. The performance of these estimation methods is evaluated through Monte Carlo simulations. Finally, the proposed approach is applied to a real dataset.