This paper uses appropriately modified information criteria to select models from the GARCH family, which are
subsequently used for predicting US dollar exchange rate return volatility. The out of sample forecast accuracy of models
chosen in this manner compares favourably on mean absolute error grounds, although less favourably on mean squared error
grounds, with those generated by the commonly used GARCH(1, 1) model. An examination of the orders of models selected
by the criteria reveals that (1, 1) models are typically selected less than 20% of the time. Ó 1998 Elsevier Science S.A. All
rights reserved.