Inference for an exponentiated half logistic distribution with application to cancer hybrid censored data

In this paper, based on hybrid censored sample from a two parameter exponentiated half logistic distribution, we consider the problem of estimating the unknown parameters using frequentist and Bayesian approaches. Expectation-Maximization, Lindley's approximation and Metropolis-Hastings algorithms are used for obtaining point estimators and corresponding confidence intervals for the shape and scale parameters involved in the underlying model. Data analyses involving the survival times of patients suffering from cancer diseases and treated radiotherapy and/or chemotherapy have been performed. Finally, numerical simulation study was conducted to assess the performances of the so developed methods and conclusions on our findings are reported.

Automated summary

In this paper, estimation of unknown parameters using frequentist and Bayesian approaches from a hybrid censored sample of a two parameter exponentiated half logistic distribution is considered. Various algorithms were used to obtain point estimators and confidence intervals for the shape and scale parameters, and data analyses on cancer patients' survival times were conducted. A numerical simulation study was carried out to evaluate the developed methods and conclusions on the findings were reported.