Modelling International Stock Market Contagion Using Copula and Risk Appetite
48 Pages Posted: 25 Oct 2007
Date Written: October 15, 2007
Abstract
Forbes and Rigobon (2002) claim there was no contagion among international stock markets during the 1997 Asian crisis, with contagion being defined as an increase in dependence. We revisit this issue using a more robust methodology based on copula. After controlling for heteroskedasticity with the skewed-t AR-GARCH model, we findnd clear evidence of contagion using dummy t-copula and two versions of time-varying t-copula. For the month from October 17, 1997 to November 17, 1997, contagion was widespread among the Asian and Europe countries, but not the Latin American ones. Contagion was also confirmed by our world and regional risk appetite indices compiled from 36 stock market indices and 32 bond market indices. Our risk appetite indices have also successfully captured five other extreme events in the last decade, including two terrorist attacks, corporate scandal in the U.S., LCTM default and the internet bubble burst. We find that the impact of terrorist attacks is small and short term, whereas the impacts of internet bubble burst and corporate scandals were more prolong and severe. Developed financial markets are the most prone to financial crisis and contagion, whereas the Latin American markets are almost isolated from all extreme events.
Keywords: Contagion, Copula, Risk Appetite
JEL Classification: G10, G15
Suggested Citation: Suggested Citation
Do you have negative results from your research you’d like to share?
Recommended Papers
-
On the Out-of-Sample Importance of Skewness and Asymmetric Dependence for Asset Allocation
-
Modelling Time-Varying Exchange Rate Dependence Using the Conditional Copula
-
Estimation of Copula Models for Time Series of Possibly Different Lengths
-
By Teng-suan Ho, Richard C. Stapleton, ...
-
A General Approach to Integrated Risk Management with Skewed, Fat-Tailed Risk
-
A General Approach to Integrated Risk Management with Skewed, Fat-Tailed Risk