Additive outliers, GARCH and forecasting volatility

The Generalized Autoregressive Conditional Heteroskedasticity [GARCH] model is often used for forecasting stock market volatility. It is frequently fo und, however, that estimated residuals from GARCH models have excess kurtos is, even when one allows for conditional t-distributed errors. In this pape r we examine if this feature can be due to neglected additive outliers [AOs ], where we focus on the out-of-sample forecasting properties of GARCH mode ls for AO-corrected returns. We find that models for AO-corrected data yiel d substantial improvement over GARCH and GARCH-t models for the original re turns, and that this improvement holds for various samples, two forecast ev aluation criteria and four stock markets. (C) 1999 Elsevier Science B.V. Al l rights reserved.