Conditional dependence between international stock markets: A long memory GARCH-copula model approach

In this paper, we investigate the relationship between major international stock markets by taking into account the long memory in volatility under structural shifts. We use long memory GARCH-skewed student-t models for the marginal distribution modeling and copulas functions for the dependence structure investigation. Using daily international stock market data from 2003 to 2017, the empirical results show that the long memory GARCH-copula models are more appropriate than standard GARCH-copulas models in dependence modeling. Moreover, results indicate that the dependence structure increases during the global financial and European debt crisis. Furthermore, a Value-at-Risk application shows that the long memory GARCH-copula models provide more accurate multivariate market risk estimation. Therefore, the dependence structure between stock markets is affected by long memory in volatility. These findings have important implications for investors interested in international stock markets for portfolio diversification, risk management, and international asset allocation.