An empirical investigation of information transmission between advanced and emerging equity markets and implication for international diversification

This dissertation investigates the impact of information transmission from Advanced Stock Markets (ASM) to Emerging Stock Markets (ESM) and the implication for international diversification. This study explores the benefits to investors in the advanced countries for holding stocks from any of the eight emerging stock markets. The Unit Roots tests by Dickey-Fuller (1981) and Kwiatkowski, Phillips, Schmidt, and Shin (KPSS 1991) are used. These techniques are used to determine the stationarity of the stock prices and the stock returns. The findings indicate that all stock prices are I(1), suggesting that the first difference of each stock price is I(0). The Augmented Dickey-Fuller (ADF) cointegration and Error Correction Models by Engle and Granger (1987) for bivariate cointegration analysis were used to investigate long-run relationships between advanced and emerging stocks, and among emerging stocks. In addition, the multivariate cointegration analyses by Johansen (1988, 1991) and Hansen and Juselius (1995) were used to examine any long-run relationships that may exist among different blocs of advanced and emerging stocks. The results indicate that some of the pairwise correlation between ASM and ESM are low, meaning that international diversification may be beneficial. Also, there is evidence of information transmission from ASM to ESM. While the Mature Emerging Stock Markets (MESM: Korea, Taiwan, and Thailand) group is influenced by Japan, the Bloc 1 portfolio has one long-run relationship. In Bloc 2, Japan/Taiwan are the cointegrated stock markets, while in Bloc 1, USA, UK and Malaysia are the source of the long-run relationship. The Common Stochastic Trends (CST) shared by all cointegrated stock prices are then extracted and the CST are responsible for the long-run relationship. Cointegration does not imply perfect correction; rather, it is only when stocks are perfectly correlated that diversification will not be beneficial. Evidence of cointegration suggest that the stock markets share a long-run equilibrium relation, and in the short-run, they are free to move apart, thereby providing opportunities for arbitrage. We conclude that stocks that are not cointegrated may be more beneficial for diversification.