Crude oil price forecasting based on a novel hybrid long memory GARCH-M and wavelet analysis model

This paper proposes a novel hybrid forecast model to forecast crude oil price on considering the long memory, asymmetric, heavy-tail distribution, nonlinear and non-stationary characteristics of crude oil price. First, we use a signal de-noising method to reduce excessive noise significantly in the crude oil price. Then we employ empirical mode decomposition to transform the de-noised price into different intrinsic mode functions (IMFs). Finally, some complex long memory GARCH-M models are used to forecast different IMFs and a residual. Empirical results show that the proposed hybrid forecasting model WPD-EMD-ARMA-FIGARCH-M achieves significant effect during periods of extreme incidents. The robustness test shows that this hybrid model is superior to traditional models. (C) 2020 Elsevier B.V. All rights reserved.