Predicting Returns on Canadian Exchange Rates With rtifical Neural Networks and Egarch-M Models

摘要

This study investigates the problem of predicting daily returns based on five Candaian exchange rates using feedforward neural networks and EGARCH models. The statistical properties of five daily exchange rate series (US dollar, German mark, French franc, Japanese yen, and British pound) are anlayzed, EGARCH-M models on the Generalized Error Distribution (GED) are fitted to the return series, and serve as comparison standard, along with random walk modes. Backpropagation networks (BPN) using lagged returns as inputs are trained and tested. Estimated volatilities from the EGARCH-M models are used also to see if aerformance is affected. The question of spillovers in interrelated markets is investigated with networks of multiple inputs and outputs. In addiiton, Elamn-tpe recurrent networks are also trained and tested. Comparison of the various methods suggests that, depsite their simplicity, neural netowrks are similar to the EGARCH-M class of nonlinear models. but superior to random walk models, in terms of in-sample fit and out-of-smaple prediction performance.