Tail dependence in a Multivariate Regular Variation framework: A Note on Financial Contagion from Russia to Latin America

The present paper examines the common wisdom on the contagion from Russia to Brazil. We concentrate our attention principally on the dependence and contagion between Brazil and Russia in the late 1998, and the potential contagion from Brazil to other countries of Latin America. We analyse the three key markets, e.g. JPM EMBI bonds, stock market index and foreign exchange rates of these countries over the daily period of 1997-2002. We introduce a nonparametric estimator for tail dependence in the constant conditional correlation GARCH framework. In this line of research we propose an empirical measure of extreme dependence to understand the joint extremal behaviour of multivariate time series, where we belive that the crisis of the nancial market would occur. Finally, our analysis does not nd a true contagion from Russia to Brazil.

Farzanegan, Mohammad Reza, Rodrigo Herrera and Alexander Karmann (2008), Tail dependence in a Multivariate Regular Variation framework: A Note on Financial Contagion from Russia to Latin America, Dresden