Asymmetric effects and long memory in dynamic volatility relationships between stock returns and exchange rates

► GARCH-class models with Student-t distributions are used to examine the volatility relationships between stock returns and exchange rates. ► Strong evidence of asymmetric reaction to news and long memory in the conditional volatility processes is found. ► Univariate FIAPARCH and bivariate CCC-FIAPARCH models provide more accurate volatility estimates and forecasts than the other competing specifications. ► FIAPARCH model is particularly useful in measuring and forecasting market risk exposure for diversified stock and currency portfolios.