Hybrid model for oil futures price forecasting

作者：Zhu Jinrong^*; Zhu Donghua; He Yongxiu; Li Furong

来源：Journal of Information and Computational Science, 2007, 4(4): 1127-1132.

• Autoregressive integrated moving average (ARIMA)
• Hybrid model
• Oil futures price

The volatility of the oil futures price is extremely complex with its nonlinear, high noise and the affiect of irregular events. Therefore, an accurate forecasting on oil futures price is an important and challenging topic. In this study, a hybrid model is developed by means of a systematic integration of autoregressive integrated moving average (ARIMA), support vector machines (SVMs) and irregular events analysis. In the proposed method, oil futures price change is decomposed into linear component, nonlinear component and events'; influence. The simulation shows that the hybrid model achieves significant improvement in the forecasting performance in comparison with the individual SVMs model and the individual ARIMA model.

• 出版日期2007
• 单位华北电力大学