Predictive analytics of the copper spot price by utilizing complex network and artificial neural network techniques

Wang, C., Zhang, X., Wang, M., Lim, M.K. and Ghadimi, P. (2019) Predictive analytics of the copper spot price by utilizing complex network and artificial neural network techniques. Resources Policy, 63, 101414. (doi: 10.1016/j.resourpol.2019.101414)

Abstract

The copper prices in international trade markets are volatile. An accurate copper price prediction may guide commodity trading and firm profits in the copper industry. In this paper, a hybrid predictive technique combining complex network and traditional artificial neural network (ANN) techniques is developed for copper price forecasting. This technique first transforms the original price time series to a price volatility network (PVN) and extracts the volatility characteristics from the topological structure of the PVN. After the original data are reconstructed, three widely used ANN techniques, the BPNN, RBFNN, and ELM, are applied to forecast the future copper price. To examine the forecasting performance of the proposed PVN-ANN techniques, the published data of copper spot prices from the New York Commodity Exchange (COMEX) are used. The empirical results show that the proposed hybrid PVN-ANN techniques can obtain a favorable prediction effect in both level and directional predictions compared to those of the traditional ANN techniques. This result clearly demonstrates the effectiveness of the proposed hybrid predictive techniques in revealing the underlying nonlinear patterns of international copper prices.

• Item Type: Articles
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Lim, Professor Ming
• Authors: Wang, C., Zhang, X., Wang, M., Lim, M.K., and Ghadimi, P.
• College/School: College of Social Sciences > Adam Smith Business School > Management
• Journal Name: Resources Policy
• Publisher: Elsevier
• ISSN: 0301-4207
• ISSN (Online): 1873-7641
• ISBN: 03014207
• Published Online: 06 June 2019