Abstract
Using realized volatility to estimate conditional variance of financial returns, we compare forecasts of volatility from linear GARCH models with asymmetric ones. We consider horizons extending to 30 days. Forecasts are compared using three different evaluation tests. With data from an equity index and two foreign exchange returns, we show that asymmetric models provide statistically significant forecast improvements upon the GARCH model for two of the datasets and improve forecasts for all datasets by means of forecasts combinations. These results extend to about 10 days in the future, beyond which the forecasts are statistically inseparable from each other.
Acknowledgements
I thank John W. Galbraith, Paul Kupiec, Matthew Thomas Jones and seminar participants at an IMF Seminar for comments. This article is based on the fourth chapter of my McGill dissertation. Data were provided by the Toronto Stock Exchange, Market Data Services and Olsen Associates, Switzerland. The views expressed herein are those of the author and should not be attributed to the IMF, its Executive Board, or its management.
Notes
1 In fact, unreported p-values show that they are generally significant at the 1% level.