Claim Missing Document
Check
Articles

Found 1 Documents
Search

A Combination of ARIMA Models and Neural Networks in Forecasting Nigerian Exchange Rate Oluwaseun Johnson Olawale; Daniel Jacob Adashu
African Multidisciplinary Journal of Sciences and Artificial Intelligence Vol 1 No 1 (2024): African Multidisciplinary Journal of Sciences and Artificial Intelligence
Publisher : Darul Yasin Al Sys

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58578/amjsai.v1i1.3367

Abstract

Over the years, the United States Dollar, European Euro, and the British Pound Sterling exchange rate to Nigerian Naira has been on the increase. It has become pertinent to identify robust models that will help to cope with the variability associated with the increase in exchange rate. Several studies showed the ARIMA method to be highly useful in modelling and forecasting exchange rates. However, not much work has been done on modelling and forecasting Nigerian Exchange rate using machine learning models which is the focus for this study. The models we used in the study are; the Autoregressive Integrated Moving Average (ARIMA), Artificial Neural Network (ANN), and ARIMA-ANN models. Secondary data obtained from Central Bank of Nigeria (CBN) were used. The results showed that the most appropriate model out of the three time series models considered for these exchange rates is the ARIMA-ANN which produced a better forecast compared to ARIMA and ANN. This conclusion was based on the lowest standards of prediction accuracy which ARIMA-ANN produced the lowest Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE) values for the three different currencies we compared against the naira. Based on the continuous increase in the Nigerian Exchange Rate, the Government and policymakers should take economic measures to avoid the persistent downfall of the Nigerian Naira.