Abstract
This article uses the global cases of COVID-19 and nine types of Morgan Stanley Capital International (MCSI) stock indexes to test the contagion effect and the impact duration of COVID-19 on the stock market between Jan 23 – Apr 14, 2020. It uses the bivariate dynamic conditional correlation generalized autoregressive conditional heteroskedasticity (DCC-GARCH) model to estimate the DCC coefficient and uses the forward forecasting test on the DCC to test the contagion effect. The empirical results show that COVID-19 has a continuous contagion effect on the ASEAN, emerging, Far-East, European with Middle Eastern, and Nordic markets, and no contagion effect on the World Index, developed, and North American markets. However, the contagion effect disappears after March 12–23.