This paper introduces a new approach to measure financial contagion in emerging markets, focusing on the co-incidence of extreme return shocks across countries within and across regions. The authors use a multinomial logistic regression model to characterize the extent, economic significance, and determinants of contagion. They find that contagion is predictable and influenced by regional interest rates, exchange rate changes, and conditional stock return volatility. The study also reveals that extreme negative returns are more contagious than extreme positive returns, but the evidence is mixed. The analysis is based on daily returns of emerging markets during the 1990s, including the 1997 Asian crisis and the 1998 Russian crisis. The authors conclude that contagion differs across regions, with Latin America showing stronger contagion within its region compared to Asia. They also find evidence of cross-regional contagion, particularly from Asia to Latin America, but not from the U.S. to other regions. The paper's approach is innovative in its use of extreme returns and multinomial logistic regression, providing a more nuanced understanding of financial contagion.This paper introduces a new approach to measure financial contagion in emerging markets, focusing on the co-incidence of extreme return shocks across countries within and across regions. The authors use a multinomial logistic regression model to characterize the extent, economic significance, and determinants of contagion. They find that contagion is predictable and influenced by regional interest rates, exchange rate changes, and conditional stock return volatility. The study also reveals that extreme negative returns are more contagious than extreme positive returns, but the evidence is mixed. The analysis is based on daily returns of emerging markets during the 1990s, including the 1997 Asian crisis and the 1998 Russian crisis. The authors conclude that contagion differs across regions, with Latin America showing stronger contagion within its region compared to Asia. They also find evidence of cross-regional contagion, particularly from Asia to Latin America, but not from the U.S. to other regions. The paper's approach is innovative in its use of extreme returns and multinomial logistic regression, providing a more nuanced understanding of financial contagion.