Working Paper Series no. 95: What is the Best Approach to Measure the Interdependence between Different Markets?

Abstract

In order to measure the interdependence between different markets, we investigate and compare different measures of dependence including cross-correlation, conditional correlation, concordance and correlation in tails. In the latter case, we use the notion of copula and we define two kinds of diagnoses which enable us to adjust the joint empirical tail distribution in the case of two or three markets for the best copulas. In particular, this approach makes it possible to understand the evolution of the interdependence of more than two markets in the tails, in particular, when extremal values (which correspond to a shock) induce some turmoil in the evolution of the markets.

Sanvi Avouyi-Dovi, Dominique Guégan and Sophie Ladoucette
December 2002

Classification JEL : C14 - C22 - G15.

Keywords : interdependence, conditional correlation, concordance, functions copulas.