Investing in Emerging Markets : An Asymmetric Stochastic Multivariate Volatility Model to Diversify Efficiently a Portfolio of Assets

The asset allocation is a practical problem for most institutional and private investors, who routinely deal with a wide variety of stocks, bonds and options. Evidence suggests that both the expected return and the volatility vary over time. Many studies find that the expected returns have time-varying components that are partially predictable, and considerable effort has been devoted to the modelling of time-varying volatility. Recent attention has moved to examining the partial implications of these properties on problems that are of practical importance. Unfortunately, initial attempts fail either to explain extreme stock market behaviour or invite significant challenge of computation.This paper investigates the daily stock market volatility in a sample of significant emerging stock markets using an Asymetric Volatility Model (ASV) model. It is estimated with the Markov Chain Monte Carlo (MCMC) method that allows the first two moments of returns to vary over time incorporating a Gibbs sampling to estimate posterior returns.Our empirical research captures most of the volatility dynamics. Interestingly, we show that both a significant persistency and the variability of volatility are higher in emerging countries as compared to advanced economies.We also provide evidence for significant negative correlation between shocks to the stock market index and shocks to volatility. The so-called “leverage effect” is tested with a sample of recent daily stock index returns throughout the period 2004-2008. Our estimation results confirm that the persistency in volatility and the variability of volatility are negatively related. Moreover, persistency in volatility and the magnitude of leverage effect are negatively correlated.We conclude by suggesting new development for the emerging country risk exposure applicable to mean variance investors. We also give some economic significance of our findings all in indicating some remaining issues for further researches