A2041
Title: Combining extreme value theory with martingale regression in market risk analytics and portfolio management
Authors: Wei Dai - Shenzhen Institute of Information Technology (China) [presenting]
Tze Leung Lai - Stanford University (United States)
Abstract: A new econometric approach is introduced to the estimation of VaR and convex risk measures in financial risk management, which uses a martingale regression for asset pricing and the extreme value theory (EVT) for the martingale difference residuals. Using decoupling inequalities, we show that the MLGEV (maximum likelihood for generalized extreme value distributions) and the peaks over threshold (POT) method in EVT can apply to the dynamically scaled residuals. This approach, therefore, addresses the pitfalls of EVT techniques while enabling them to realize their widely recognized potential for estimating extreme quantiles and probabilities.
