Menu Content/Inhalt
Seminars Print
previous year previous month next month next year
See by year See by month See by week See Today Search Jump to month
David Veredas, ECARES Print
Monday, 12 December 2011, 12:00 - 13:30

David Veredas, ECARES

A New Measure of Tail Correlation for Vast Dimensional Panels of Asset Returns

Abstract: The new keyword of the forthcoming Solvency II accords is "tail correlation". Though its importance is empirically demonstrated, its measurement is not straightforward. We introduce TailCor, a simple measure for tail correlation for vast dimensional panels of asset returns. It is quantile-based (and therefore moment free) and we only require the returns to follow an elliptical distribution. It differs from copula-based tail dependence and extreme value theory tail measures in that TailCor is not based on tail asymptotic arguments, and hence exact estimation can be done for any probability level. The use of TailCor is straightforward: it is a simple function and it disentangles the contribution of the linear and non-linear correlations, the latter depending on the tail index. A throughout Monte Carlo study shows the goodness of the method, both in terms of computational time and for finite samples. An empirical illustration to a panel of securities (14 insurance and 36 banks, all Europeans) over Nov. 2001 to Nov. 2011 illustrates the usefulness of TailCor.

Location: CORE-LLN
Contact: Sylvie Mauroy, This e-mail address is being protected from spam bots, you need JavaScript enabled to view it