Many contemporary studies assume that, as the economies become more closely integrated and cross-border financial flows accelerate, national capital markets become more highly correlated. Moreover, the IMF Economic Outlook 2000 and the Economist 2001 indicated that because of rapid growth in global fund transfers, cross-border activities of financial flow, international divisions in the electronic industry, and deregulations, the global market epidemic effect has been significantly increasing, especially during excitable periods. The traditional single-β ICAPM doesn’t consider the relationships between the market volatilities and the market correlations, and then assumes constant risk coefficient of individual country’s stock asset. In this paper, we use the SWARCH models to identify the specific market volatility states at each time point, and then analyze the cross-market correlations, the multi-β ICAPM settings and the abnormal returns from the various volatility combinations from the individual and world stock markets. Our results are consistent with the following notions. First, the greatest returns correlations exist when both individual and world markets are at the high volatility states. Second, the maximum (minimum) β coincides with the individual and world markets at the high and low (low and high) volatility states, respectively. Third, the differential β settings from various combinations of volatility states may be one of drivers to the documented abnormal returns.