Bayesian and non-bayesian inference of the generalized Lomax distribution under a unified hybrid censoring scheme with applications in failure times in biomedical and aerospace materials

The unified hybrid censoring scheme is a combination of different types of censoring schemes used in reliability testing. This paper presents the statistical inference of generalized Lomax distribution under unified hybrid censoring scheme. The point and interval estimates of the parameters $$\alpha , \beta $$ α , β , and $$\gamma $$ γ of the generalized Lomax distribution have been studied for unified hybrid censored data. In point estimation, the maximum likelihood estimation method is used for computing the estimates, and Tierney and Kadane estimation method is used for Bayes estimation. A $$100 (1-\sigma )\%$$ 100 ( 1 - σ ) % approximate confidence interval and Bayesian credible intervals for the parameters $$\alpha , \beta $$ α , β , and $$\gamma $$ γ have been computed in the interval estimation part. Mean squared errors are computed for all the estimates and comparison of estimates have been done. The results indicate that the Bayesian estimation method yields more accurate and reliable parameter estimates compared to the maximum likelihood approach. Finally, data representing failure times of fatigue fracture of Kevlar 373/epoxy and failure times of aircraft windshields have been used for point and interval estimations of all parameters as application of real-life scenarios.