PERBANDINGAN METODE MOMEN, MAXIMUM LIKELIHOOD DAN BAYES DALAM MENDUGA PARAMETER DISTRIBUSI PARETO

A. Nurul Amalia (Prodi Statistika, FMIPA, Universitas Negeri Makassar)
Muhammad Arif Tiro (Prodi Statistika, FMIPA, Universitas Negeri Makassar)
Aswi Aswi (Prodi Statistika, FMIPA, Universitas Negeri Makassar)

Variansi : Journal of Statistics and Its Application on Teaching and Research
Vol 3, No 3 (2021)

This study examines the estimation of Pareto distribution parameters using three different methods, namely the Moment, Maximum Likelihood, and Bayesian methods. The Pareto distribution is a continuous distribution with parameters k > 0 and α > 0. These two parameters are estimated by using three distinct parameter estimation methods. The goodness of fit measure used in choosing the best estimation method is the Mean Square Error (MSE) value. The smallest MSE is the best method. A simulation study is carried out as well as the case study of the data on the number of Gross National Income (GNI) per capita in Southeast Asian countries in 2019. The estimation and simulation results indicate that the best estimation method in estimating the parameters of the Pareto distribution is the Maximum Likelihood in terms of MSE value.Keywords: Pareto distribution, Moment Method, Maximum Likelihood IMethod, Bayesian Method