News and Correlations of CEEC-3 Financial Markets

Abstract

We investigate conditional correlations between six CEEC-3 financial markets estimated by DCC-MGARCH models. In general, the highest correlations exist between Hungary and Poland in foreign exchange and stock markets. Short-term money markets are rather isolated from each other. We find that the associations of CEEC-3 exchange rates versus the euro are weaker than those versus the US dollar. The persistence of the effect of shocks on the timevarying correlations is strongest for foreign exchange and stock markets, indicating a tendency toward contagion. In searching for the origins of financial market volatility in the CEEC-3, we uncover some evidence of Granger-causality on the foreign exchange markets. Finally, using a pool model, we investigate the impact of euro area, US, and CEEC-3 news on the correlations. Apart from ECB monetary policy news, we observe no broad effects of international news on correlations; instead, local news exerts an influence, which suggests a dominance of country- or market-specific circumstances.