Issue

Vol. 3 No. 1 (2026): January

Copyright (c) 2026 Mohammad Mansour Ataey; Abdul Kabir Azizi (Author)

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Forecasting Gold Price Volatility Using Econometric and Machine-Learning Models

https://doi.org/10.62810/jssh.v3i1.231
Mohammad Mansour Ataey
Balkh University, Master’s Student in Finance and Accounting, Faculty of Economics, Mazar-i-Sharif, AF
https://orcid.org/0009-0009-5990-2088
Abdul Kabir Azizi
Balkh University, Department of Finance and banking, Faculty of Economics, Mazar-i-Sharif, AF
https://orcid.org/0000-0001-7833-2409

Corresponding Author(s) : Mohammad Mansour Ataey

mansourataey@gmail.com

Journal of Social Sciences and Humanities, Vol. 3 No. 1 (2026): January

Abstract

This paper presents a critical comparative analysis of classic econometric models and current machine-learning methods for predicting the daily volatility of the gold price. As a safe-haven asset used globally, Gold exhibits significant nonlinear dynamics, structural breaks, and consistent volatility clustering, making it difficult to predict accurately, which is necessary for both investors and policymakers. Interestingly, even with much research, there are still many gaps in long-horizon datasets, integrated comparative frameworks, and out-of-sample assessment of econometric and machine-learning models. To seal these gaps, this research uses daily XAU/USD data from 2010-2024 and analyzes the forecasting results of ARIMA, GARCH(1,1), and data mining (Random Forest and XGBoost) models. The analysis is performed using a Python-based empirical framework that includes data preprocessing, feature engineering, diagnostics for stationarity and heteroskedasticity, and performance evaluation using MAE, RMSE, MAPE, and R2. The results indicate that the leptokurtic and stationary characteristics of gold returns, along with high volatility concentration, limit the predictive power of linear econometric models. Machine-learning models significantly outperform ARIMA and GARCH, with the XGBoost model providing the best results across all measures of accuracy. These findings underscore the advantages of nonlinear, data-driven models for volatility regime changes and have beneficial implications for traders, portfolio managers, and agencies of financial stability seeking more dependable volatility forecasting instruments.

Keywords

ARIMA Gold volatility GARCH Machine learning XGBoost
Ataey, M. M., & Azizi, A. K. (2026). Forecasting Gold Price Volatility Using Econometric and Machine-Learning Models. Journal of Social Sciences and Humanities, 3(1), 173–188. https://doi.org/10.62810/jssh.v3i1.231
Download Citation
Endnote/Zotero/Mendeley (RIS)
BibTeX
References
  1. Baur, D. G., & Lucey, B. M. (2020). Is gold a hedge or a safe haven? An analysis of stocks, bonds, and gold. Financial Review, 55(1), 1–27. https://doi.org/10.1111/j.1540-6288.2010.00244.x DOI: https://doi.org/10.1111/j.1540-6288.2010.00244.x
  2. Bergmeir, C., & Benítez, J. M. (2012). On the use of cross-validation for time series predictor evaluation. Information Sciences, 191, 192–213. https://doi.org/10.1016/j.ins.2011.12.028 DOI: https://doi.org/10.1016/j.ins.2011.12.028
  3. Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics, 31(3), 307–327. https://doi.org/10.1016/0304-4076(86)90063-1 DOI: https://doi.org/10.1016/0304-4076(86)90063-1
  4. Box, G. E. P., & Jenkins, G. M. (1976). Time series analysis: Forecasting and control. Holden-Day.
  5. Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324 DOI: https://doi.org/10.1023/A:1010933404324
  6. Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785–794). https://doi.org/10.1145/2939672.2939785 DOI: https://doi.org/10.1145/2939672.2939785
  7. Dickey, D. A., & Fuller, W. A. (1979). Distribution of the estimators for autoregressive time series with a unit root. Journal of the American Statistical Association, 74(366), 427–431. https://doi.org/10.2307/2286348 DOI: https://doi.org/10.1080/01621459.1979.10482531
  8. Engle, R. F. (1982). Autoregressive conditional heteroscedasticity with estimates of the variance of UK inflation. Econometrica, 50(4), 987–1007. https://doi.org/10.2307/1912773 DOI: https://doi.org/10.2307/1912773
  9. Engle, R. F. (2021). Financial volatility and risk management. Annual Review of Financial Economics, 13, 1–24. https://doi.org/10.1093/jjfinec/nbaa038 DOI: https://doi.org/10.1093/jjfinec/nbaa038
  10. Feng, X., He, J., & Chen, S. (2022). Machine learning for financial market prediction: A survey. Expert Systems with Applications, 198, 116804.
  11. Feng, X., Li, Q., & Wang, Z. (2022). Financial volatility forecasting with machine learning: A comprehensive review. Finance Research Letters, 48, 102937.
  12. Granger, C. W., & Poon, S. H. (2001). Forecasting financial market volatility: A review. Available at SSRN 268866.
  13. Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: Principles and practice (2nd ed.). OTexts. DOI: https://doi.org/10.32614/CRAN.package.fpp2
  14. Hyndman, R. J., & Koehler, A. B. (2006). Another look at measures of forecast accuracy. International Journal of Forecasting, 22(4), 679–688. https://doi.org/10.1016/j.ijforecast.2006.03.001 DOI: https://doi.org/10.1016/j.ijforecast.2006.03.001
  15. Kumari, S. N., & Tan, A. (2018). Modeling and forecasting volatility series: with reference to gold price. Thailand Statistician, 16(1), 77-63.
  16. Li, J., Wang, R., Aizhan, D., & Karimzade, M. (2023). Assessing the impacts of Covid-19 on stock exchange, gold prices, and financial markets: Fresh evidence from econometric analysis. Resources Policy, 83, 103617. https://doi.org/10.1016/j.resourpol.2023.103617 DOI: https://doi.org/10.1016/j.resourpol.2023.103617
  17. Li, Y., Jia, N., Yu, X., Manning, N., Lan, X., & Liu, J. (2023). Transboundary flows in the metacoupled Anthropocene: typology, methods, and governance for global sustainability. Ecology and Society, 28(3). https://doi.org/10.5751/ES-14351-280319 DOI: https://doi.org/10.5751/ES-14351-280319
  18. Poon, S.-H., & Granger, C. W. J. (2003). Forecasting volatility in financial markets: A review. Journal of Economic Literature, 41(2), 478–539. https://doi.org/10.1257/002205103765762743 DOI: https://doi.org/10.1257/jel.41.2.478
  19. Rundo, F., Trenta, F., & Di Buono, M. V. (2019). Machine learning for financial applications: A survey. Applied Sciences, 9(24), 5574. https://doi.org/10.3390/app9245574 DOI: https://doi.org/10.3390/app9245574
  20. Tsay, R. S. (2010). Analysis of financial time series (3rd ed.). Wiley. DOI: https://doi.org/10.1002/9780470644560
  21. Zhang, Y., Li, X., & Wu, Y. (2023). Gold market volatility and global uncertainty: Evidence from advanced models. Economic Modelling, 122, 106355. https://doi.org/10.3390/toxics12070526 DOI: https://doi.org/10.3390/toxics12070526
  22. Zare, M. (2025). Forecasting market returns using machine learning: evidence from Random Forest models. Applied Economics Letters, 1-5. https://doi.org/10.1080/13504851.2025.2567614 DOI: https://doi.org/10.1080/13504851.2025.2567614
Read More
Author biographies is not available.
Received 2025-11-19
Accepted 2026-01-26
Published 2026-01-31
Download
PDF
Statistic
Read Counter : 126 Download : 32