Autoregressive Fractional Integrated Moving Average (ARFIMA(p,d,q)) Modelling of Nigeria Exchange

The Autoregressive Fractionally Integrated Moving Average (ARFIMA) Model was used to model the Nigerian currency rate in this study. The Augmented Dickey Fuller (ADF) test was used to perform the unit root test at both the level and fractional difference. Stationarity was achieved in the fractional difference series (0.0868). The Geweke and Porter-Hudak (GPH) approach was used to estimate the ARFIMA model's long memory parameter d. The sample autocorrelation function demonstrated the presence of a long memory structure. The ARFIMA (1, 0.0868, 1) models were chosen as the best model based on minimal information criteria (AIC values).

Please see the link :- https://globalpresshub.com/index.php/AJPAM/article/view/1422