Resumen de Inference based on censored samples from exponentiated populations

Essam K. AL-Hussaini

• Maximum likelihood and Bayes estimators of the parameters, survival and hazard rate functions of an exponentiated population whose cumulative distribution function takes the form H(x)=[G(x)] α , where α is an unknown positive parameter and G(x) is a cumulative distribution function, are obtained in two cases: when the parameters of G(x) are all known so that α is the only unknown parameter of H and when the k-dimensional vector of parameters β of G(x) are all unknown so that the (k+1)-dimensional vector of parameters (α,β) of H is unknown. In both cases, estimation is based on a Type II censored sample. In such cases, the predictive density and predictive survival functions of future observables are obtained when based on the two-sample scheme. This enables us to obtain Bayes prediction bounds of future observables. Subjective priors are used which reflect the prior belief of the experimenter. Numerical results are obtained under three exponentiated models and compared with the results obtained by using the MCMC algorithm.