A Hybrid Intelligent System for Forecasting Gasoline Price. A Comparative Approximate Economic

The difficulty in gasoline price forecasting has attracted muchattention of academic researchers and business practitioners.Various methods have been tried to solve the problem of forecastinggasoline prices however, all of the existing models of prediction cannotmeet practical needs. In this paper, a novel hybrid intelligent frameworkis developed by applying a systematic integration of GMDH neuralnetworks with GA and Rule-based Exert System (RES) with Web-basedText Mining (WTM) employs for gasoline price forecasting. Ourresearch reveals that during the recent financial crisis period byemploying hybrid intelligent framework for gasoline price forecasting,we obtain better forecasting results compared to the GMDH neuralnetworks and results will be so better when we employ hybrid intelligentsystem with GARCH (1, 1) for gasoline price volatility forecasting