Search

In this paper, we discuss the inference for the competing risks model when the failure times follow Chen distribution. With assumption of two causes of failures, which are partially observed, are considered as independent. The existence and uniqueness of maximum likelihood estimates for model parameters are obtained under generalized progressive hybrid censoring. Also, we discussed the classical and Bayesian inferences of the model parameters under the assumption of restricted and nonrestricted parameters. Performance of classical point and interval estimators are compared with Bayesian point and interval estimators by conducting extensive simulation study. In addition to that, for illustration purpose, a real life example is discussed. Finally, some concluding remarks, regarding the presented model, are made. [ABSTRACT FROM AUTHOR]

Inference for a Gumbel Type-II model is considered under hybrid Type-I censoring. The EM algorithm is considered for maximum likelihood estimation and approximate confidence intervals are evaluated using the Fisher information matrix. The Bayesian inferences are obtained with respect to different loss functions assuming weighted Lindley priors and corresponding intervals are discussed. Further, predictive estimates of censored observations are studied. Optimal censoring schemes are calculated based on different information measure criteria. Studied inferential methods are compared from extensive Monte Carlo simulations. Finally, the methods are illustrated with the analysis of a real data set. [ABSTRACT FROM AUTHOR]

The accelerated life testing (ALT) is an efficient approach and has been used in several fields to obtain failure time data of test units in a much shorter time than testing at normal operating conditions. In this article, a progressive‐stress ALT under progressive type‐II censoring is considered when the lifetime of test units follows logistic exponential distribution. We assume that the scale parameter of the distribution satisfying the inverse power law. First, the maximum likelihood estimates of the model parameters and their approximate confidence intervals are obtained. Next, we obtain Bayes estimators under squared error loss function with the help of Metropolis‐Hasting (MH) algorithm. We also derive highest posterior density (HPD) credible intervals of the model parameters. Monte Carlo simulations are performed to compare the performances of the proposed methods of estimation. Finally, one data set has been analyzed for illustrative purposes. [ABSTRACT FROM AUTHOR]