Groupe d’études et de recherche en analyse des décisions

G-2000-02

Learning and Forecasting Foreign Exchange Rates using Recurrent Neural Network Dynamics

et

Changes in financial market are collective human decisions which are result of the dynamics of the neurons. In this paper, we describe a discrete time Recurrent Neural Network model for the exchange rate changes of some major currencies. We have been able to show that this model after going through a proper learning process with some historical data, is capable of tracking those foreign exchange rate trajectories quite closely.

, 13 pages