Chocs, chocs de volatilité et contagion entre les marchés boursiers. Application d'un modèle icss-mgarch

In this paper we study dynamic interdependence between stock exchange markets across indices returns especially during crisis periods. For this purpose, we analyze the behaviour of eleven stock exchange markets from Europe, North America and Asia during the period 1985-2007. We use a Vector Autoregressive Multivariate garch model with multiple regime in variance. Breakpoints in variance<np pagenum="278"/> are given by Iterated Cumulative Sum of Squared algorithm corrected for heteroskedasticity. Our results on the transmission phenomena proof the significant mean and variance causality from us to all markets. Also, we find no evidence of shocks transmission from emerging markets towards to the developed markets one. Concerning the markets cross-correlation, the dynamic analysis enables us to check the asymmetrical pattern and its instability. Finally, the cross- correlation during crisis periods indicates a significant rise particularly for the American crash of 1987. There is a contagion from us market to all markets. However, contagion is not observed frequently for others crisis. At last but not the least, openness is not a tool to isolate markets from instability and contagion. Markets need regulation and supervision even they could be perfect.garchusus Classification JEL : F36 ; C32 ; G15