Semi-parametric estimation of joint large movements of risky assets

The classical approach to modelling the occurrence of joint large movements of asset returns is to assume multivariate normality for the distribution of asset returns. This implies independence between large returns. However, it is now recognised by both academics and practitioners that large movements of assets returns do not occur independently. This fact encourages the modelling joint large movements of asset returns as non-normal, a non trivial task mainly due to the natural scarcity of such extreme events. This paper shows how to estimate the probability of joint large movements of asset prices using a semi-parametric approach borrowed from extreme value theory (EVT). It helps to understand the contribution of individual assets to large portfolio losses in terms of joint large movements. The advantages of this approach are that it does not require the assumption of a specific parametric form for the dependence structure of the joint large movements, avoiding the model misspecification; it addresses specifically the scarcity of data which is a problem for the reliable fitting of fully parametric models; and it is applicable to portfolios of many assets: there is no dimension explosion. The paper includes an empirical analysis of international equity data showing how to implement semi-parametric EVT modelling and how to exploit its strengths to help understand the probability of joint large movements. We estimate the probability of joint large losses in a portfolio composed of the FTSE 100, Nikkei 250 and S&P 500 indices. Each of the index returns is found to be heavy tailed. The S&P 500 index has a much stronger effect on large portfolio losses than the FTSE 100, although having similar univariate tail heaviness.

[1]  L. Haan,et al.  Residual Life Time at Great Age , 1974 .

[2]  J. Pickands.  Statistical Inference Using Extreme Order Statistics , 1975 .

[3]  B. M. Hill,et al.  A Simple General Approach to Inference About the Tail of a Distribution , 1975 .

[4]  S. Resnick,et al.  Limit theory for multivariate sample extremes , 1977 .

[5]  R. Engle.  Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation , 1982 .

[6]  Dennis W. Jansen,et al.  On the Frequency of Large Stock Returns: Putting Booms and Busts into Perspective , 1989 .

[7]  L. Haan,et al.  A moment estimator for the index of an extreme-value distribution , 1989 .

[8]  Sidney I. Resnick,et al.  Estimating the limit distribution of multivariate extremes , 1993 .

[9]  F. Longin.  The Asymptotic Distribution of Extreme Stock Market Returns , 1996 .

[10]  François M. Longia.  Value at Risk and Extreme Values , 1998 .

[11]  Laurens de Haan,et al.  Sea and Wind: Multivariate Extremes at Work , 1998 .

[12]  F. Diebold,et al.  Pitfalls and Opportunities in the Use of Extreme Value Theory in Risk Management , 1998 .

[13]  Ser-Huang Poon,et al.  Returns synchronization and daily correlation dynamics between international stock markets , 1999 .

[14]  C. Stărică,et al.  Multivariate extremes for models with constant conditional correlations , 1999 .

[15]  F. Longin,et al.  Extreme Correlation of International Equity Markets , 2000 .

[16]  Stephen E. Satchell,et al.  Evaluating the Performance of Nearest Neighbour Algorithms when Forecasting US Industry Returns , 2000 .

[17]  Anthony D. Hall,et al.  Using Bayesian Variable Selection Methods to Choose Style Factors in Global Stock Return Models , 2000 .

[18]  Jon Danielsson,et al.  Value-at-Risk and Extreme Returns , 2000 .

[19]  A. McNeil,et al.  Estimation of tail-related risk measures for heteroscedastic financial time series: an extreme value approach , 2000 .

[20]  J. Corcoran.  Modelling Extremal Events for Insurance and Finance , 2002 .

[21]  J. Tawn,et al.  Extreme Value Dependence in Financial Markets: Diagnostics, Models, and Financial Implications , 2004 .

[22]  P. Hartmann,et al.  Asset Market Linkages in Crisis Periods , 2001, Review of Economics and Statistics.

[23]  F. Longin,et al.  The choice of the distribution of asset returns: How extreme value theory can help? , 2005 .

[24]  Extreme Value Estimation of Boom and Crash Statistics , 2006 .

[25]  S. Resnick.  Heavy-Tail Phenomena: Probabilistic and Statistical Modeling , 2006 .

[26]  Simone Alfarano,et al.  A Simple Asymmetric Herding Model to Distinguish Between Stock and Foreign Exchange Markets , 2007 .

[27]  Thomas Lux,et al.  Empirical validation of stochastic models of interacting agents , 2007 .

[28]  James M. Nason,et al.  Exchange Rates and Fundamentals: A Generalization , 2008 .