Crude Oil Price Modeling: a Multi Scale Wavelet Neural Network Perspective
Crude Oil Price Modeling: a Multi Scale Wavelet Neural Network Perspective

Author(s): Rania Jammazi, Chaker Aloui
Subject(s): Economy
Published by: ASERS Publishing
Keywords: Harr à Trous wavelet; neural network; back propagation; crude oil price forecasting; activation function; input-hidden nodes
Summary/Abstract: Oil price prediction has usually proved to be an intractable task, due to the intrinsic complexity of the oil market mechanism. In addition, the recent oil shock and its consequences re-launch the debate on understanding the behavior underlying the expected oil prices. Combining the dynamic properties of multilayer back propagation neural network and the recent Harr à Trous wavelet decomposition, a Hybrid model HTW-MPNN is implemented to achieve prominent modeling of crude oil price. While recent studies focus on the determination of the best model by comparing various neural architectures or applying several decomposition techniques to the ANN, the new insight of this chapteris to target the issue of the transfer function selection. Based on the work of authors Yonaba H. Anctil F. and Fortin, V., from the year 2010 “Comparing Sigmoid Transfer Functions for Neural Network Multistep Ahead Stream flow forecasting”(published in the Journal of Hydrologic Engineering, April, 275-283), we use three variants of activations function namely sigmoid, bipolar sigmoid and hyperbolic tangent in order to test the model’s flexibility. Furthermore, the forecasting robustness is checked through several levels of input-hidden nodes. Comparatively, simulation results of Hybrid model performs better than the conventional BPNN.

• Details
• Contents