Volatility Prediction of Shanghai Gold Futures Based on Transfer Learning_Vol. 4 No. 1 (JBDC 2026)_Journal of Big Data and Computing (ISSN: 2959-0590)

Home > Journal of Big Data and Computing (ISSN: 2959-0590) > Vol. 4 No. 1 (JBDC 2026) >

Volatility Prediction of Shanghai Gold Futures Based on Transfer Learning

DOI: https://doi.org/10.62517/jbdc.202601130

Author(s)

Jingyun Zhao*

Affiliation(s)

School of Mathematics and Statistics, North China University of Water Resources and Electric Power, Zhengzhou, Henan, China *Corresponding Author

Abstract

Gold futures volatility forecasting is of great significance for asset allocation, risk management, and the interpretation of macroeconomic signals. However, its nonlinearity, long memory, and cross- market heterogeneity pose challenges to both traditional econometric models and deep learning methods. This paper proposes a transfer learning-based volatility forecasting model for futures, named CrossFormer-GRU, which transfers knowledge from the Brent crude oil market to the Shanghai gold futures market. By incorporating an uncertainty-weighted adversarial domain adaptation mechanism, the model dynamically optimizes multi-task losses. Experimental results show that CrossFormer-GRU significantly outperforms benchmark models such as GARCH, SVR, LSTM, and Transformer in terms of MAE, RMSE, MAPE, and R². Ablation studies further validate the critical roles of transfer learning and uncertainty weighting in enhancing model performance.

Keywords

Gold Futures; Transfer Learning; Uncertainty Weighting; CrossFormer-GRU

References

[1]LI Bin, WANG Wen, ZHANG Jingze. US Stock Market Shocks and the Link Between Gold Futures Prices in China and the US. Financial Market Research, 2024, (12): 120-129. [2]ZHONG M R, SHI Y, YIN L B, et al. Research on the Impact of Investor Attention on Gold Futures Market from the Perspective of High Frequency. Journal of Systems Science and Mathematical Sciences, 2020, 40(11): 1935- 1949. [3]ZHONG Meirui, SHI Yue, YIN Libo, et al. Research on the Impact of Investor Attention on Gold Futures Market from the Perspective of High Frequency. Journal of Systems Science and Mathematical Sciences, 2020, 40(11): 1935- 1949. [4]Wang L, Wang Z, Qu H, et al. Optimal forecast combination based on neural networks for time series forecasting. Applied Soft Computing, 2018, 66: 1-17. [5]Ampountolas A. Enhancing Forecasting Accuracy in Commodity and Financial Markets: Insights from GARCH and SVR Models. International Journal of Financial Studies, 2024, 12(3): 59-59. [6]Bentes RS. Forecasting volatility in gold returns under the GARCH, IGARCH and FIGARCH frameworks: New evidence. Physical A: Statistical Mechanics and its Applications, 2015, 438355-364. [7]Yu W, Yi zhi W, M.BL, et al. Cryptocurrency uncertainty and volatility forecasting of precious metal futures markets. Journal of Commodity Markets, 2023, 29 [8]Chen Y, Ye N, Zhang W, et al. Meta-LSTR: Meta-Learning with Long Short-Term Transformer for futures volatility prediction. Expert Systems With Applications, 2025, 265125926- 125926. [9]Kim Christensen, Mathias Siggaard, Bezirgen Veliyev. A machine learning approach to volatility forecasting. Financial econometrics, 2025, 23, 032, https://doi.org/10.1093/jjfinec/nbac032 [10]Fischer T, Krauss C. Deep Learning with Long Short-Term Memory Networks for Financial Market Predictions. European Journal of Operational Research, 2018 (270): 654-669. [11]A. Vaswani, N. Shazeer, N. Parmar, et al. Attention is all you need, in Advances in Neural Information Processing Systems 30 (NeurIPS 2017), Long Beach, CA, USA, Dec. 2017, 5998-6008. [12]H. Zhou, S. Zhang, J. Peng, et al. Informer: Beyond efficient transformer for long sequence time-series forecasting, in Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), vol. 35, no. 12, 2021, pp. 11106-11115. [13]LAPTEV N. Reconstruction and Regression Loss for Time-Series Transfer Learning//24th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. London: ACM, 2018: 1-8. [14]YE R, DAI Q.A Relationship-Aligned Transfer Learning Algorithm for Time Series Forecasting. Information Sciences, 2022, 593:17-34. [15]NGUYEN T T, YOON S. A Novel Approach to Short-Term Stock Price Movement Prediction Using Transfer Learning. Applied Sciences, 2023, 9(22):4745 (2019-11-07). [16]Kendall A, Gal Y, Cipolla R. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018: 7482-7491 [17]Wang R, Zhang R, Henao R. Wasserstein uncertainty estimation for adversarial domain matching. Frontiers in Big Data, 2022, 5: 878716.