Forecasting the exchange rate of the Algerian dinar against the US dollar using an Autoregressive Model, Integrated Moving Averages, and Artificial Neural Networks – a comparative study

Abstract

This research paper aims to evaluate and compare the predictive performance of two models for forecasting the Algerian dinar exchange rate against the US dollar: the Autoregressive Integrated Moving Average (ARIMA) model, and the Artificial Neural Network (NAR-ANN) model, To achieve this objective, monthly data (01/2000 to 12/2025) was used. This data was divided into two sets: a training set covering the first 281 months for estimating the two models, and a test set covering the last 31 months for testing and comparing them. The applied results, comparing the predictive performance of the two models using prediction accuracy criteria, showed that the NAR-ANN model outperformed the ARIMA model in predicting the Algerian dinar exchange rate against the US dollar.

Jel Classification Codes:F31,E47