|
Activity Number:
|
39
- Methods in Financial Risk Assessment
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 30, 2017 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Risk Analysis
|
|
Abstract #323118
|
|
|
Title:
|
Modeling Maxima in Financial Time Series with Dynamic Generalized Extreme Value Distribution
|
|
Author(s):
|
Zifeng Zhao* and Zhengjun Zhang and Rong Chen
|
|
Companies:
|
University of Wisconsin-Madison and University of Wisconsin and Rutgers University
|
|
Keywords:
|
Extreme value theory ;
Dynamic modeling ;
Financial risk management ;
High-frequency data ;
Value at risk ;
Generalized extreme value distribution
|
|
Abstract:
|
This paper integrates the static generalized extreme value(GEV) distribution with dynamic modeling approach to introduce a novel dynamic GEV framework for the modeling of maxima in financial time series. An autoregressive dynamic GEV(dGEV) model is proposed which allows for time-varying scale parameter(volatility) and shape parameter(tail index) of a Type-II GEV(Frechet) distribution. dGEV provides a direct and accurate modeling of the time varying behavior of maxima and offers a new angle to study the tail risk dynamics in financial markets. Probabilistic properties of dGEV are fully studied and an irregular maximum likelihood estimator is used for model estimation, with its statistical properties investigated. Simulation study shows the flexibility of dGEV and confirms the reliability of its estimators. The results of two real data examples in which dGEV is used for market tail risk monitoring and VaR calculation are presented, where signi ficant improvement over static GEV has been observed. Empirical result of dGEV is consistent with the findings of the dynamic peak-over-threshold(POT) literature, that the tail index of fi nancial markets varies through time.
|
Authors who are presenting talks have a * after their name.
