Semi-Parametric Estimation of Joint Large Movements of Risky Assets
25 Pages Posted: 23 Mar 2009
Date Written: March 18, 2009
Abstract
This article presents a semi-parametric framework for estimating the probability of joint large movements of asset prices using extreme value theory. The advantages of this approach are: no parametric form for the dependence structure of the joint large movements has to be specified, avoiding the model misspecification; it addresses specifically the scarcity of data, a problem when fitting fully parametric models; it is suitable for portfolios of many assets, there is no dimension explosion.
We show the strength of the approach using returns on three international equity indices: S&P500, Nikkei250 and FTSE100. We find that although the S&P500 and FTSE100 have similar univariate tail heaviness, the former makes a much larger contribution to the probability of large losses on an equally weighted portfolio.
Keywords: portfolio tail probability,risk management,multivariate extreme value theory
JEL Classification: C51,G11
Suggested Citation: Suggested Citation