Estimation of parameters in erlang distribution using prior information

Abstract

This paper focuses on the estimation of parameters of an Erlang density using known Coefficients of variation and Kurtosis of the population basing on past experience and a simple random sample of size n from the population. Estimators using Searle D.T approach are proposed and their bias B(T) and mean squared error M(T) are calculated. The relative efficiency of T over conventional estimator x is also tabulated for various sample sizes and various C.V values, Kurtosis values. The proposed estimators are observed to be more efficient than x under the conditions established. Subject Classification: AMS 62J13.