Value-at-risk forecasting based on the Asymmetric Exponential Power distribution
Open Access
Author:
Ou, Lu
Graduate Program:
Statistics
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
July 08, 2016
Committee Members:
Zhibiao Zhao, Thesis Advisor/Co-Advisor
Keywords:
Asymmetric Exponential Power distribution Value-at-risk Forecasting
Abstract:
Value-at-risk (VaR) is a standard measure of market risk in financial markets. This paper proposes a novel, adaptive, and flexible method to forecast volatility, asymmetry, and VaR. As an extension of the existing exponential smoothing as well as GARCH formulations, the method is motivated from an Asymmetric Exponential Power distribution, which includes the Laplace, Normal, and Uniform distributions as special cases and takes into account the potentially time-varying nature in volatility and skewness of financial return series. Results from a Monte Carlo simulation study and an empirical application to the S&P 500 index illustrate that the proposed method offers more accurate and robust VaR forecasts across a range of models at different confidence levels compared to other popular parametric alternatives.
