|
Activity Number:
|
339
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Risk Analysis
|
| Abstract - #309144 |
|
Title:
|
Extreme Value Theorem and Its Applications
|
|
Author(s):
|
Yan Li*+
|
|
Companies:
|
The Pennsylvania State University
|
|
Address:
|
Statistics Department, 265 Blue Course Dr, State College, PA, 16803,
|
|
Keywords:
|
EVT ; Operational risk ; POT
|
|
Abstract:
|
Extreme risks are the concern of risk managers and investors. In order to estimate and hedge the risks better, Extreme Value Theory (EVT) has been fully developed in the past fifty years. The modeling framework based on the operational risk scenario will be established. The heavily skewed loss data histogram suggests a fat-tailed distribution such as the lognormal or Pareto distribution. Besides those standard parametric distributions, more advanced technique focusing on tail behavior, EVT, provides a better approach to fit the skewed data. Two principals of EVT, Block Maxima Models and Peaks-over-Thresholds Model, will be presented. The important asymptotic theories and the estimation procedures will be briefly introduced as well.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2007 program |
