Ensemble Regression-Based Gold Price (XAU/USD) Prediction

Abstract

The prediction of any commodities such as cryptocurrency, stocks, silver, gold is a challenging task for the investors, researchers, and analysts. In this work, we propose a model that forecasts the value of 1 ounce of gold in dollars by using regression ensemble-based approaches. To our knowledge, this is the very first study in terms of combining regression models for the prediction of XAU/USD index although there are plenty of methods employed in the literature to forecast the price of gold. The contributions of this study are fivefold. First, the dataset is gathered between July 2019 and July 2020 from global financial websites in the world, and cleaned for modeling. Then, feature space is extended with technical and statistical indicators in addition to opening, closing, highest, lowest prices of gold index. Next, different regression and ensemble-based regression models are carried out. These are linear regression, polynomial regression, decision tree regression, random forest regression, support vector regression, voting regressor, stacking regressor. Experiment results demonstrate that the usage of stacking regression combination model exhibits considerable results with 2.2036 of MAPE for forecasting the price of XAU/USD index.

Keywords

• Gold price prediction
• XAU/USD index forecast
• ensemble regression
• stacking regressor

References

1. Ismail, Z., Yahya, A., & Shabri, A. (2009). Forecasting gold prices using multiple linear regression method. American Journal of Applied Sciences, 6(8), 1509.
2. Manoj, J., & Suresh, K. K. (2019). Forecast Model for Price of Gold: Multiple Linear Regression with Principal Component Analysis. Thailand Statistician, 17(1), 125-131.
3. Jianwei, E., Ye, J., & Jin, H. (2019). A novel hybrid model on the prediction of time series and its application for the gold price analysis and forecasting. Physica A: Statistical Mechanics and Its Applications, 527, 121454.
4. Wei, Y., Liang, C., Li, Y., Zhang, X., & Wei, G. (2020). Can CBOE gold and silver implied volatility help to forecast gold futures volatility in China? Evidence based on HAR and Ridge regression models. Finance Research Letters, 35, 101287.
5. Pierdzioch, C., Risse, M., & Rohloff, S. (2015). A real-time quantile-regression approach to forecasting gold returns under asymmetric loss. Resources Policy, 45, 299-306.
6. Suranart, K., Kiattisin, S., & Leelasantitham, A. (2014, March). Analysis of comparisons for forecasting gold price using neural network, radial basis function network and support vector regression. In The 4th Joint International Conference on Information and Communication Technology, Electronic and Electrical Engineering (JICTEE) (pp. 1-5). IEEE.
7. Ongsritrakul, P., & Soonthornphisaj, N. (2003, July). Apply decision tree and support vector regression to predict the gold price. In Proceedings of the International Joint Conference on Neural Networks, 2003. (Vol. 4, pp. 2488-2492). IEEE.
8. Sadorsky, P. (2021). Predicting gold and silver price direction using tree-based classifiers. Journal of Risk and Financial Management, 14(5), 198.

9. M. A. Mithu, K. M. Rahman, R. A. Razu, M. Riajuliislam, S. I. Momo and A. Sattar. (2021, July). Gold price forecasting using regression techniques for settling economic and stock market inconsistency. In 12th International Conference on Computing Communication and Networking Technologies (ICCCNT), pp. 1-4, IEEE.
10. Yanto, M., Sanjaya, S., Yulasmi, Y., Guswandi, D., & Arlis, S. (2021). Implementation multiple linear regresion in neural network predict gold price. Indonesian Journal of Electrical Engineering and Computer Science, 22(3), 1635-1642.
11. Freedman, D. A. (2009). Statistical models: theory and practice. cambridge university press.
12. Rencher, A. C., & Christensen, W. F. (2012). Chapter 10, Multivariate regression–Section 10.1, Introduction. Methods of multivariate analysis, Wiley Series in Probability and Statistics, 709, 19.
13. Hilary, L. (1967). Seal. Studies in the history of probability and statistics. XV: The historical development of the Gauss linear model. Biometrika, 1-24.
14. Yan, X., & Su, X. (2009). Linear regression analysis: theory and computing. World Scientific.
15. Ostertagová, E. (2012). Modelling using polynomial regression. Procedia Engineering, 48, 500-506.
16. Quinlan, J. R. (1990). Decision trees and decision-making. IEEE Transactions on Systems, Man, and Cybernetics, 20(2), 339-346.
17. Wu, X., Kumar, V., Ross Quinlan, J., Ghosh, J., Yang, Q., Motoda, H., McLachlan, G. J., Angus, L. B., Yu, P. S., Zhou, Z., Steinberg, D. (2008). Top 10 algorithms in data mining. Knowledge and information systems, 14(1), 1-37.
18. Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32.
19. Korting, T. S. (2006). C4. 5 algorithm and multivariate decision trees. Image Processing Division, National Institute for Space Research–INPE Sao Jose dos Campos–SP, Brazil, 22.
20. Vapnik, V. N. (1999). An overview of statistical learning theory. IEEE transactions on neural networks, 10(5), 988-999.
21. Vapnik, V., Golowich, S., & Smola, A. (1996). Support vector method for function approximation, regression estimation and signal processing. Advances in neural information processing systems, 9.
22. An, K., & Meng, J. (2010, August). Voting-averaged combination method for regressor ensemble. In International Conference on Intelligent Computing (pp. 540-546). Springer, Berlin, Heidelberg.
23. Ruta, D., & Gabrys, B. (2005). Classifier selection for majority voting. Information fusion, 6(1), 63-81.
24. Džeroski, S., & Ženko, B. (2004). Is combining classifiers with stacking better than selecting the best one?. Machine learning, 54(3), 255-273.