Forecasting Exchange-Rates via Local Approximation Methods and Neural Networks

There has been an increased number of papers in the literature in recent years, applying several methods and techniques for exchange-rate prediction. This paper focuses on the Greek drachma using daily observations of the drachma rates against four major currencies, namely the U.S. Dollar (USD), the Deutsche Mark (DM), the French Franc (FF) and the British Pound (GBP) for a period of 11 years, aiming at forecasting their short-term course by applying local approximation methods based on both chaotic analysis and neural networks.