Title: Gold price forecasting with a neuro-fuzzy-based inference system

Authors: Georgia Makridou; George S. Atsalakis; Constantinos Zopounidis; Kostas Andriosopoulos

Addresses: Research Centre for Energy Management, ESCP Europe Business School, UK; Technical University of Crete, Crete, Greece ' Technical University of Crete, Department of Production Engineering and Management, Financial Engineering Laboratory, University Campus, 73100 Chania, Greece ' Technical University of Crete, Department of Production Engineering and Management, Financial Engineering Laboratory, University Campus, 73100 Chania, Greece ' Research Centre for Energy Management, ESCP Europe Business School, UK

Abstract: Following the importance of gold in the global economy and the high interest that has attracted recently, the objective of this paper is twofold: to predict the price of gold by using the Adaptive Neuro-Fuzzy Inference System (ANFIS) and compare its forecasting accuracy with various time-series forecasting methods and the 'Buy and Hold' (B&H) strategy. The results show that the ANFIS's accuracy is far superior to the performance of all compared methods and therefore ANFIS demonstrates the potential of neuro fuzzy-based modelling for predicting the gold's price.

Keywords: gold price forecasting; adaptive neuro-fuzzy inference systems; ANFIS; B&H strategy; buy and hold; time series forecasting; financial management; neural networks; fuzzy logic; forecasting accuracy; modelling.

DOI: 10.1504/IJFERM.2013.053707

International Journal of Financial Engineering and Risk Management, 2013 Vol.1 No.1, pp.35 - 54

Received: 06 Sep 2012
Accepted: 15 Oct 2012
Published online: 10 Sep 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article