Community Research and Development Information Service - CORDIS


This paper analyzes bivariate and multivariate market association with a modified cross-sectional correlation measure based on Solnik and Roulet (2000). The behaviour of this measure is evaluated for different simulated correlation processes and volatility characteristics in a bivariate and multivariate setting. It is shown that the measure allows a more fundamental analysis of the sources of time-varying market association than the correlation coefficient. An empirical analysis of eleven developed stock markets investigates the evolution of the market association and its volatility through time. It can be shown that the global market association has increased if all markets are analyzed simultaneously. However, this result cannot be confirmed for all pairs of markets in a bivariate setting. In addition, asymmetric effects of positive and negative shocks can only be found for some pairs of markets and not for all markets simultaneously.

Additional information

Authors: BAUR D, European Commission, Joint Research Council, Institute for the Protection and the Security of the Citizen, Ispra (IT)
Bibliographic Reference: EUR 20701 EN (2003), 28pp. Free of charge
Availability: Available from European Commission, JRC Knowledge Management Unit, Ispra (IT) Tel: +39 033278 9843 or +39 033278 9864 Fax: +39 033278 9623 E-mail:
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top