A New Mixture Pareto Distribution and Its Application

Authors

  • Nareerat Nanuwong Department of Statistics, Faculty of Science, Kasetsart University
  • Winai Bodhisuwan Department of Statistics, Faculty of Science, Kasetsart University
  • Chookait Pudprommarat

Keywords:

length biased, two-component mixture, mixture Pareto distribution, maximum likelihood estimation, lifetime data, hazard rate

Abstract

This paper introduces a new mixture Pareto distribution generated from logit of the weighted two-component mixture distribution which is mixed between a Pareto and a length biased Pareto distributions. Special sub-models include the Pareto, exponential, chi-square and logistic distributions. We contain derivation of the mixture Pareto types I,II , III and IV provide various mathematical properties including its limit behavior, hazard rate and  rth moment. Discussion of the estimation procedure is based on maximum likelihood. Also, we display an application to Norwegian fire claim data, it shows that the mixture Pareto distribution give a better fit than other important lifetime models; a Weibull, the Pareto and the length biased Pareto distributions. In conclusion, the mixture Pareto distribution provides a rather general and flexible framework for the statistical lifetime data analysis.

Downloads

Published

2015-07-30

How to Cite

Nanuwong, N., Bodhisuwan, W., & Pudprommarat, C. (2015). A New Mixture Pareto Distribution and Its Application. Thailand Statistician, 13(2), 191–207. Retrieved from https://ph02.tci-thaijo.org/index.php/thaistat/article/view/37772

Issue

Section

Articles