The Performance of Arima and Arfima in Modelling the Exchange Rate of Nigeria Currency to Other Currencies

Economic performance of a nation depends majorly on the stability of foreign exchange rate; the economic viability hangs on the exchange rate of local currencies against other currencies across the globe. Box – Jenkins Approach was employed to model the Naira exchange rate to other major currencies using Autoregressive Integrated Moving Average (ARIMA) and The autoregressive fractional integral moving average (ARFIMA) models. This studies aimed on measuring forecast ability of Autoregressive Integrated Moving Average (ARIMA) (p,d,q) and autoregressive fractional integral moving average (ARFIMA) (p, fd, q) models for stationary type series that exhibit features of Long memory properties. Results indicate autoregressive fractional integral moving average (ARFIMA) is the best model in terms of fit, serial correlation analysis and accuracy measures. The out-sample forecasts confirmed the competence of the autoregressive fractional integral moving average (ARFIMA) models as shown by forecast validation tools. Consequently, the out-sample forecasts result nearly reveal the current economic situation in Nigeria indicating that the autoregressive fractional integral moving average (ARFIMA) model is appropriate and realistic in modeling and forecasting the strength of Naira to other currencies.