Forecasting crude oil prices volatility by reconstructing EEMD components using ARIMA and FFNN models

The energy sector which includes gas and oil is concerned to explore develop refined it’s a multitrillion business. As crude very important source of energy, it has valuable impact on country’s economic growth, national security, social stability. Therefore, accurately predicting the price volatility topic research still, challenge for researchers forecast prices. this study conducted address said problem significantly. This presents novel hybrid method reconstructing EEMD IMFs that involves two steps. Visual analysis Average Mutual Information (AMI) graphs were used rebuild IMFs. split into components called stochastic deterministic. In proposed method, reconstruction was done at stages see if have more variation. Later, ARIMA FFNN models test suggested method’s performance. For purpose, Brent prices data used, model EEMD-S2D1D2-ARIMA/FFNN outperformed other existing with minimum MAE = 0.2323, RMSE 0.3058 MAPE 0.5273. A simulation also check robustness N 50, 500, 1,000, 2000, 5,000, 7,500. results confirm unpredictability present in reconstructed EEMD-ARIMA/FFNN EEMD-SD-ARIMA/FFNN been reduced by models.