NONLINEAR FORECASTING OF THE GOLD MINER SPREAD: AN APPLICATION OF CORRELATION FILTERS
Article first published online: 30 SEP 2013
Copyright © 2013 John Wiley & Sons, Ltd.
Intelligent Systems in Accounting, Finance and Management
Volume 20, Issue 4, pages 207–231, October/December 2013
How to Cite
Dunis, C. L., Laws, J., Middleton, P. W. and Karathanasopoulos, A. (2013), NONLINEAR FORECASTING OF THE GOLD MINER SPREAD: AN APPLICATION OF CORRELATION FILTERS. Int. J. Intell. Syst. Acc. Fin. Mgmt., 20: 207–231. doi: 10.1002/isaf.1345
- Issue published online: 3 DEC 2013
- Article first published online: 30 SEP 2013
- Manuscript Revised: 10 AUG 2013
- Manuscript Accepted: 10 AUG 2013
- Manuscript Received: 28 AUG 2012
- spread trading;
- multilayer perceptron neural network;
- particle swarm optimization;
- radial basis function neural network;
- genetic programming algorithm;
- correlation filter
This paper models and forecasts the Gold Miner Spread from 23 May 2006 to 30 June 2011. The Gold Miner Spread acts as a suitable performance indicator for the relationship between physical gold and US gold equity.
The contribution of this investigation is twofold. First, the accuracy of each model is evaluated from a statistical perspective. Second, various forecasting methodologies are then applied to trade the spread. Trading models include an ARMA (12,12) model, a cointegration model, a multilayer perceptron neural network (NN), a particle swarm optimization radial basis function NN and a genetic programming algorithm (GPA).
Results obtained from an out-of-sample trading simulation validate the in-sample back test as the GPA model produced the highest risk-adjusted returns. Correlation filters are also applied to enhance performance and, as a consequence, volatility is reduced by 5%, on average, while returns are improved between 2.54% and 8.11% across five of the six models. Copyright © 2013 John Wiley & Sons, Ltd.