A Nonparametric Method for Extreme Values

• Activity Number: 309
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Computing
• Author(s): Mei Ling Huang*+ and Lucas Thorpe
• Companies: Brock University and Brock University
• Keywords: Extreme value ; heavy tailed distribution ; goodness of fit test ; order statistics ; Pareto distribution

Abstract

Pareto distribution is a popular distribution with many applications in the real world extreme events. Many existing estimation methods for Pareto distribution are based on estimating the tail index. In this paper we propose a nonparametric distribution estimation method and study its properties. The results of Monte Carlo simulations show good properties of the proposed method. The paper also studies several examples of real world applications on extreme values by using proposed method. We compare the results on goodness-of-fit tests by applying the proposed nonparametric method and existing parametric methods with real-world data.