Equity market interdependence: the relationship between European and US stock markets

In this article, the degree of interdependence between European and US stock markets is measured by the conditional correlation between stock returns: the correlation coefficient is estimated using a model describing the variations over time in a number of variables (returns and volatility, for example), and its estimate takes account of all available information at a given time. We estimate conditional variance in the same way. Moreover, two statistical tools, recently introduced in applied finance, are combined. The first, developed by Engle in 2001 – an original specification of the conditional correlations in multivariate models – enables us to describe time-varying correlations between two or more assets. The second tool, copula functions, allows us to apply distributions that are more consistent with the stylised facts observed on financial markets than those commonly used. The approach used in this study is original in that it combines both the above tools. Using a multivariate model implies rejecting the two assumptions traditionally adopted in empirical studies in finance: correlations between assets are presumed to be constant; asymmetry or the presence of rare events are not taken into account in asset price distributions. Consequently, our empirical findings corroborate the assumption that correlations vary over time and validate the choice of an asymmetric joint distribution integrating the presence of rare events. We also observe the presence of periods of strong and weak correlations and similar periods for volatility. Furthermore, our results highlight a close link between the correlations and volatilities observed on the different equity markets: in phases of high volatility, the correlation tends to rise above its medium-term average; inversely, in phases of low volatility, markets seem to display greater independence. Lastly, the correlation coefficient of close to 1 confirms that French and German stock market indices have been converging in recent years. This may reflect the growing integration of these two markets and of the economies of these two countries within Economic and Monetary Union.