CMStatistics 2017: Start Registration
View Submission - CMStatistics
Title: Characterizing the concentration of tail risk under fat-tails Authors:  Pasquale Cirillo - Delft University of Technology (Netherlands) [presenting]
Abstract: The aim is to discuss how concentration measures can be used to study fat-tails, when dealing with the size distributions of income, wealth, losses, etc. Focusing on the Gini index, we will go beyond its use as a concentration measure and will show how to define it as a probabilistic measure of risk, and in particular of tail risk, deriving a completely brand new set of tools for the study of fat-tails. The goal is to better characterize tail risk in the case of a very large or infinite population variance. We will naturally explain the main theoretical aspects and properties, but we will also discuss heuristics and applications on actual data.