Forecasting Daily Exchange Rates with Artificial Neural Network (ANN)

Abstract

The modeling and forecasting of nominal exchange rate dynamics has long been a focus of financial and economic studies. Artificial intelligence (AI) modeling has recently received a great deal of attention as a new method in economic and financial forecasting. This research suggests an alternate strategy for forecasting daily exchange rates that is based on artificial neural network (ANN).Our empirical research is based on a set of Tunisian daily data. In order to evaluate this strategy, we compare its performance to that of a generalized autoregressive conditional heteroskedasticity (GARCH) model. The results show that the proposed nonlinear autoregressive (NAR) model is a reliable and fast prediction method. This discovery allows businesses and policymakers to plan more effectively.