Department of Mathematics and Statistics, School of Natural Sciences and Mathematics, Chinhoyi University of Technology, 78 Magamba Way, Off Chirundu Road, Chinhoyi, Zimbabwe.
World Journal of Advanced Engineering Technology and Sciences, 2025, 17(03), 481–490
Article DOI: 10.30574/wjaets.2025.17.3.1579
Received on 12 November 2025; revised on 29 December 2025; accepted on 31 December 2025
No time series modelling strategy performs consistently better than others in all situations. Different methods yield differing efficacies for different scenarios. This study compared the nonlinear autoregressive neural network (NARNN) and seasonal autoregressive integrated moving average (SARIMA) methods in modelling the United States Dollar – Chinese Yuan monthly average exchange rates over the period from January 2020 to November 2025. The NARNN outperformed the SARIMA and predicted consistent increases in exchange rates from December 2025 to March 2026. Regulators, speculators, policy makers and investors can make appropriate strategic decisions in anticipation of the statistically inferred fluctuations in the near future.
Nonlinear Autoregressive Neural Network (NARNN); Seasonal Autoregressive Integrated Moving Average (SARIMA); Time Series Modelling And Forecasting.
Get Your e Certificate of Publication using below link
Preview Article PDF
Henry Samambgwa and Thomas Musora. Modelling and Forecasting the United States Dollar – Chinese Yuan Exchange Rate | Nonlinear Autoregressive Neural Network vs Seasonal Autoregressive Integrated Moving Average. World Journal of Advanced Engineering Technology and Sciences, 2025, 17(03), 481-490. Article DOI: https://doi.org/10.30574/wjaets.2025.17.3.1579.