Parameter estimation methods for the Marshall-Olkin Power Lomax distribution and its applications

Abstract

The Marshall-Olkin Power Lomax (MOPLx) distribution, being extended the Power Lomax distribution, is obtained by using Marshall-Olkin (MO) scheme. This distribution can be applied for skewed data set and used as an alternative to Gamma or Weibull distributions. The main objective of this paper is to compare common method, maximum likelihood estimator (MLE) with the minimum Anderson-Darling estimator (MADE), for the shape and scale parameters of the MOPLx distribution. The efficiency of both estimators for MOPLx distribution is examined via a simulation study with different sample sizes and its specified parameters by using bias, standard deviation and root mean square error as the criteria for the comparison between MLE and MADE. Furthermore, for the applications of the MOPLx distribution, four life time data sets are exemplified to evaluate the performances of these estimators for both small and large samples.