Estimating the Parameter of a Geometric Distribution From Series System Data

Abstract

In a traditional setup of estimation of an unknown parameter of component lifetime distribution, system's continuous lifetime data is used. In this paper, we propose a simple and competitive estimator that is based on discrete lifetime data, i.e., the number of failed components at the time when the system fails. In particular, we consider the estimation of the parameter of a geometric distribution based on the system's lifetime data, and the number of failed components upon the failure of the system when the system has a series structure. Two moment estimators that are based on the system lifetime data and the number of failed components at the moment of system failure are obtained and their performances are compared in terms of the mean square error. The associated Bayesian estimators with non -informative priors are also discussed.