The Gompertz distribution has been used to describe human mortality and establish actuarial tables. Recently, this distribution has been again studied by some authors. The maximum likelihood estimates for the parameters of the Gompertz distribution has been discussed by Garg et al. [J. R. Statist. Soc. C 19 (1970) 152]. The purpose of this paper is to propose unweighted and weighted least squares estimates for parameters of the Gompertz distribution under the complete data and the first failure-censored data (series systems; see [J. Statist. Comput. Simulat. 52 (1995) 337]). A simulation study is carried out to compare the proposed estimators and the maximum likelihood estimators. Results of the simulation studies show that the performance of the weighted least squares estimators is acceptable.
Applied Mathematics and Computation 158(1), pp.133-147