Comparison of different methods of parameters estimation for Pareto Model
Journal Title: Caspian Journal of Applied Sciences Research - Year 2013, Vol 2, Issue 1
Abstract
In this study, the scale and the shape parameters of the Pareto Distribution have been estimated using five different estimation techniques, namely Method of Moments, Maximum Likelihood Estimation, Fractional Moments, Probability Weighted Moments and Bayesian method. As a single choice of sample size and parameter point do not help to clarify performance of the methods, so different parameter points and different sample sizes are used. An extensive Monte Carlo simulation study has been conducted to investigate the performance of the estimators. The WinBUGS and R-Language are used to deal with numerical computations of estimates of parameters of Pareto distribution. The Bayesian method exhibits the minimum standard error with some exceptions.
Authors and Affiliations
Rizwan Munir, Muhammad Saleem, Sajid Ali
Keywords
Related Articles
- EP ID EP130400
- DOI -
- Views 93
- Downloads 0
Forecasting Stream Flow in Floodplain by Utilizing Hydrological and Statistical Relationship in the Sepidroud Catchments
In the present study hydro-meteorological data of 5 years (2004-2008) have been used for determining stream flow in Sepidroud watersheds (North of Iran-Guilan). 4 years of hydro-meteorological data (2004-2007) were...
Design of a Statcom-Based Harmonic-Free Static VAR Compensator for Load Balancing Purposes
In this paper, a bipolar (capacitive and inductive) static VAR compensator is built on the basis of statcom fundamentals. The designed compensator exchanges wide range of pure reactive power with the ac supply with...
Performance Characteristics of Large Volume BGO for Prompt Gamma Neutron Activation AnalysisPerformance Characteristics of Large Volume BGO for Prompt Gamma Neutron Activation Analysis
A general design of an optimized prompt gamma neutron activation analysis technique is evaluated. This technique utilizes large volume BGO detector and radioisotopic neutron source. The system incorporates two isot...
Estimation and Prediction of Inverse Lomax Model via Bayesian Approach
Statistical analysis via Bayesian approach is a common practice applied to draw inferences about unknown parameter(s) of a probability distribution. This article includes Bayesian inference of unknown shape parame...
Weight-length relationships and condition factors for Sturgeon fish species from the Southeastern Caspian Sea, Iranian waters
Sturgeons are among the most endangered fish species in the worldwide. Length and weight descriptive statistics are reported for 5 sturgeon species caught off the Caspian Sea during a 2 year survey (2009-2010). Dur...