MASTER'S THESIS AT THE UNIVERSITY OF BASRAH DISCUSSES THE USE OF THE ARFIMA-RBFN HYBRID MODEL IN PREDICTION WITH APPLICATION
A master's thesis in the Department of Statistics at the College of Administration and Economics at the University of Basrah (Using the ARFIMA-RBFN hybrid model in forecasting with application).
The thesis presented by the student (Mukhtar Hussein Mukhtar Al-Essa) aims to say that the ARFIMA-RBFN hybrid model combines the predictive power of the ARFIMA model in processing linear patterns and long memory, and the ability to capture non-linear patterns by the RBFN network. This combination provides a powerful and accurate tool for predicting time series.
The thesis recommended the use of hybrid models such as ARFIMA-RBFN when analyzing complex time series, as these models can achieve a balance between representing long-term effects and non-linear relationships. It also recommended using the results of this study to improve investment decisions in the oil sector, especially with reliance on the accurate predictions of the hybrid model.