Correlation in price changes and volatility of major Latin American stock markets

This paper seeks to investigate the short-run dynamics between five major Latin American stock markets (Argentina, Brazil, Chile, Colombia and Mexico). Unlike previous research on these markets, the joint distribution of stock returns is estimated as a vector autoregression (VAR) with innovations following an exponential GARCH process. Our study is carried out using closing stock market prices covering the period 25 May 1992 to 16 May 1997. The results revealed that these countries have significant first and second moment time dependencies. In general, the markets of these countries exhibit stronger volatility than mean spillovers. Further, our results indicate that these markets exhibit stronger volatility spillovers than other regions of the world. In view of these dependencies, the conditional correlations between these markets are different from their conventional simple counterparts. Since the correlation is the catalyst in portfolio diversification and an essential parameter in the calculation of the cost of capital, we anticipate that international investors and corporate managers will find our results very interesting.