Estimation of the Weibull Parameters under Type I Censoring

Abstract

Abstract The authors propose estimators for the shape and the scale parameters in the Weibull distribution under Type I censoring. They derive exact moments of the estimator for the shape parameter and establish the joint asymptotic normality of both estimators. The proposed estimators, compared with the corresponding maximum likelihood estimators, provide a simpler procedure while maintaining high asymptotic relative efficiencies for a wide range of censoring levels. Estimation of the reliability function and the mean life and the relationship to the Type II case are also discussed.