Modified Moment Estimation for a Two Parameter Gamma Distribution

Abstract

A study of a modification of moment estimation of a two parameter gamma distribution has been carried out. The scale parameter has been estimated using the Modified Moment Estimates (MMEs) and their performance is compared to the Maximum Likelihood Estimates (MLEs). Modified Moment Estimates are easy to compute relative to Maximum Likelihood Estimates. To adjust their bias, the Jackknife technique has been used. The efficiency of these estimators is compared by calculating their respective variances. The R-has been used to carry out the Monte Carlo simulation of data. The results obtained show that the Modified Moment Estimates perform as well as the Maximum Likelihood Estimates when bias as well as variation of data is taken into account.