- Article
- Open access
- Published:
Scientific Reports (2026) Cite this article
We are providing an unedited version of this manuscript to give early access to its findings. Before final publication, the manuscript will undergo further editing. Please note there may be errors present which affect the content, and all legal disclaimers apply.
Abstract
Stock market crashes pose a significant risk to investors and regulators; they are difficult to predict in their early stages because of market volatility. The conventional machine learning methods often struggle to learn financial time series having complex structures such as non-linearity, high dimensionality, and dynamic volatility, often found in a real-life turbulent market. Although gold and silver have been considered relatively stable investment assets, the crashing of these metal rates impacts retail investors as well as global finance in a powerful manner. To overcome these limitations, the current work presents a hybrid quantum-classical Machine Learning (QCML) approach for forecasting commodity market crashes that combines Quantum Support Vector Machines (QSVM) with eXtreme Gradient Boosting (XGBoost). In order to spot crash events, we have proposed a new drawdown-based labeling approach based on significant price declines as well as short-term recovery trends. Feature engineering and training is performed on baselines such as volatility measures, momentum indicators, and trend-based ratios. The proposed design also features Random Forest thresholding, ensemble weight optimization, and feature selection. As practical quantum hardware with sufficient capability is currently unavailable for our limited experiment, the quantum component is implemented using a simulator. The proposed Hybrid QSVM–XGBoost framework achieved its best performance for the Gold 3-month prediction horizon with an ROC-AUC of 0.81. The silver market, in contrast, has lower predictability (ROC-AUC = 0.62) because of greater volatility and complex dynamics. Our results demonstrate the feasibility of integrating simulated quantum-kernel learning with classical machine learning techniques for commodity market crash prediction. While the proposed framework shows promising predictive performance, it does not claim quantum advantage or quantum supremacy under current NISQ-era hardware limitations.
Subjects
Acknowledgements
The authors acknowledge that language editing tools (Grammarly and QuillBot) are used only to assist with grammar and spelling refinement during the preparation of this study.
Funding
The authors declare that no funding was received for this research.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Reprints and permissions
About this article
Cite this article
Singh, R.D., Muhuri, S. & Singh, Y. A hybrid quantum–classical machine learning framework for commodity market crash prediction. Sci Rep (2026). https://doi.org/10.1038/s41598-026-66427-y
Download citation
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-66427-y