IDEAS home Printed from https://ideas.repec.org/p/arx/papers/2508.06497.html
   My bibliography  Save this paper

Forecasting Commodity Price Shocks Using Temporal and Semantic Fusion of Prices Signals and Agentic Generative AI Extracted Economic News

Author

Listed:
  • Mohammed-Khalil Ghali
  • Cecil Pang
  • Oscar Molina
  • Carlos Gershenson-Garcia
  • Daehan Won

Abstract

Accurate forecasting of commodity price spikes is vital for countries with limited economic buffers, where sudden increases can strain national budgets, disrupt import-reliant sectors, and undermine food and energy security. This paper introduces a hybrid forecasting framework that combines historical commodity price data with semantic signals derived from global economic news, using an agentic generative AI pipeline. The architecture integrates dual-stream Long Short-Term Memory (LSTM) networks with attention mechanisms to fuse structured time-series inputs with semantically embedded, fact-checked news summaries collected from 1960 to 2023. The model is evaluated on a 64-year dataset comprising normalized commodity price series and temporally aligned news embeddings. Results show that the proposed approach achieves a mean AUC of 0.94 and an overall accuracy of 0.91 substantially outperforming traditional baselines such as logistic regression (AUC = 0.34), random forest (AUC = 0.57), and support vector machines (AUC = 0.47). Additional ablation studies reveal that the removal of attention or dimensionality reduction leads to moderate declines in performance, while eliminating the news component causes a steep drop in AUC to 0.46, underscoring the critical value of incorporating real-world context through unstructured text. These findings demonstrate that integrating agentic generative AI with deep learning can meaningfully improve early detection of commodity price shocks, offering a practical tool for economic planning and risk mitigation in volatile market environments while saving the very high costs of operating a full generative AI agents pipeline.

Suggested Citation

  • Mohammed-Khalil Ghali & Cecil Pang & Oscar Molina & Carlos Gershenson-Garcia & Daehan Won, 2025. "Forecasting Commodity Price Shocks Using Temporal and Semantic Fusion of Prices Signals and Agentic Generative AI Extracted Economic News," Papers 2508.06497, arXiv.org.
  • Handle: RePEc:arx:papers:2508.06497
    as

    Download full text from publisher

    File URL: http://arxiv.org/pdf/2508.06497
    File Function: Latest version
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Chiroma, Haruna & Abdulkareem, Sameem & Herawan, Tutut, 2015. "Evolutionary Neural Network model for West Texas Intermediate crude oil price prediction," Applied Energy, Elsevier, vol. 142(C), pages 266-273.
    2. Zhao, Yang & Li, Jianping & Yu, Lean, 2017. "A deep learning ensemble approach for crude oil price forecasting," Energy Economics, Elsevier, vol. 66(C), pages 9-16.
    3. Yang Li & Yangyang Yu & Haohang Li & Zhi Chen & Khaldoun Khashanah, 2023. "TradingGPT: Multi-Agent System with Layered Memory and Distinct Characters for Enhanced Financial Trading Performance," Papers 2309.03736, arXiv.org.
    4. Christiane Baumeister & Lutz Kilian, 2015. "Forecasting the Real Price of Oil in a Changing World: A Forecast Combination Approach," Journal of Business & Economic Statistics, Taylor & Francis Journals, vol. 33(3), pages 338-351, July.
    5. Sánchez Lasheras, Fernando & de Cos Juez, Francisco Javier & Suárez Sánchez, Ana & Krzemień, Alicja & Riesgo Fernández, Pedro, 2015. "Forecasting the COMEX copper spot price by means of neural networks and ARIMA models," Resources Policy, Elsevier, vol. 45(C), pages 37-43.
    6. repec:fpr:gsspwp:140687 is not listed on IDEAS
    7. Paul C. Tetlock, 2007. "Giving Content to Investor Sentiment: The Role of Media in the Stock Market," Journal of Finance, American Finance Association, vol. 62(3), pages 1139-1168, June.
    8. Laborde Debucquet, David & Mamun, Abdullah, 2024. "Role of international price and domestic inflation in triggering export restrictions on food commodities," IFPRI discussion papers 2246, International Food Policy Research Institute (IFPRI).
    9. Banerjee, Ameet Kumar & Sensoy, Ahmet & Goodell, John W. & Mahapatra, Biplab, 2024. "Impact of media hype and fake news on commodity futures prices: A deep learning approach over the COVID-19 period," Finance Research Letters, Elsevier, vol. 59(C).
    10. Marek Kwas & Michał Rubaszek, 2021. "Forecasting Commodity Prices: Looking for a Benchmark," Forecasting, MDPI, vol. 3(2), pages 1-13, June.
    11. Shijie Wu & Ozan Irsoy & Steven Lu & Vadim Dabravolski & Mark Dredze & Sebastian Gehrmann & Prabhanjan Kambadur & David Rosenberg & Gideon Mann, 2023. "BloombergGPT: A Large Language Model for Finance," Papers 2303.17564, arXiv.org, revised Dec 2023.
    12. Wheeler,Collette Mari & Baffes,John & Kabundi,Alain Ntumba & Kindberg-Hanlon,Gene & Nagle,Peter Stephen Oliver & Ohnsorge,Franziska Lieselotte, 2020. "Adding Fuel to the Fire : Cheap Oil during the COVID-19 Pandemic," Policy Research Working Paper Series 9320, The World Bank.
    13. Li, Yuze & Jiang, Shangrong & Li, Xuerong & Wang, Shouyang, 2021. "The role of news sentiment in oil futures returns and volatility forecasting: Data-decomposition based deep learning approach," Energy Economics, Elsevier, vol. 95(C).
    14. Sepehr Ramyar & Farhad Kianfar, 2019. "Forecasting Crude Oil Prices: A Comparison Between Artificial Neural Networks and Vector Autoregressive Models," Computational Economics, Springer;Society for Computational Economics, vol. 53(2), pages 743-761, February.
    15. Alain de Janvry & Elisabeth Sadoulet, 2010. "Agricultural Growth and Poverty Reduction: Additional Evidence," The World Bank Research Observer, World Bank, vol. 25(1), pages 1-20, February.
    16. Abbott, Philip C. & Hurt, Christopher & Tyner, Wallace E., 2011. "What’s Driving Food Prices in 2011?," Issue Reports 112927, Farm Foundation.
    17. Qi Zhang & Yi Hu & Jianbin Jiao & Shouyang Wang, 2022. "Exploring the Trend of Commodity Prices: A Review and Bibliometric Analysis," Sustainability, MDPI, vol. 14(15), pages 1-22, August.
    18. Mamun, Abdullah & Laborde Debucquet, David, 2024. "Role of international price and domestic inflation in triggering export restrictions on food commodities," IFPRI discussion papers 2246, International Food Policy Research Institute (IFPRI).
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Haas, Christian & Budin, Constantin & d’Arcy, Anne, 2024. "How to select oil price prediction models — The effect of statistical and financial performance metrics and sentiment scores," Energy Economics, Elsevier, vol. 133(C).
    2. Xu Gong & Keqin Guan & Qiyang Chen, 2022. "The role of textual analysis in oil futures price forecasting based on machine learning approach," Journal of Futures Markets, John Wiley & Sons, Ltd., vol. 42(10), pages 1987-2017, October.
    3. Zheng, Li & Sun, Yuying & Wang, Shouyang, 2024. "A novel interval-based hybrid framework for crude oil price forecasting and trading," Energy Economics, Elsevier, vol. 130(C).
    4. Arroyo Marioli,Francisco & Khadan,Jeetendra & Ohnsorge,Franziska Lieselotte & Yamazaki,Takefumi, 2023. "Forecasting Industrial Commodity Prices : Literature Review and a Model Suite," Policy Research Working Paper Series 10611, The World Bank.
    5. Radosław Puka & Bartosz Łamasz & Marek Michalski, 2021. "Effectiveness of Artificial Neural Networks in Hedging against WTI Crude Oil Price Risk," Energies, MDPI, vol. 14(11), pages 1-26, June.
    6. Nida Çakır Melek & Charles W. Calomiris & Harry Mamaysky, 2020. "Big Data Meets the Turbulent Oil Market," Research Working Paper RWP 20-20, Federal Reserve Bank of Kansas City, revised Nov 2022.
    7. Yijia Xiao & Edward Sun & Tong Chen & Fang Wu & Di Luo & Wei Wang, 2025. "Trading-R1: Financial Trading with LLM Reasoning via Reinforcement Learning," Papers 2509.11420, arXiv.org.
    8. Jha, Nimish & Kumar Tanneru, Hemanth & Palla, Sridhar & Hussain Mafat, Iradat, 2024. "Multivariate analysis and forecasting of the crude oil prices: Part I – Classical machine learning approaches," Energy, Elsevier, vol. 296(C).
    9. Xing, Li-Min & Zhang, Yue-Jun, 2022. "Forecasting crude oil prices with shrinkage methods: Can nonconvex penalty and Huber loss help?," Energy Economics, Elsevier, vol. 110(C).
    10. Wang, Minggang & Zhao, Longfeng & Du, Ruijin & Wang, Chao & Chen, Lin & Tian, Lixin & Eugene Stanley, H., 2018. "A novel hybrid method of forecasting crude oil prices using complex network science and artificial intelligence algorithms," Applied Energy, Elsevier, vol. 220(C), pages 480-495.
    11. Chi, Yeguang & El-Jahel, Lina & Vu, Thanh, 2024. "Novel and old news sentiment in commodity futures markets," Energy Economics, Elsevier, vol. 140(C).
    12. Kumar Kulbhaskar, Anamika & Subramaniam, Sowmya, 2023. "Breaking news headlines: Impact on trading activity in the cryptocurrency market," Economic Modelling, Elsevier, vol. 126(C).
    13. Zhao, Geya & Xue, Minggao & Cheng, Li, 2023. "A new hybrid model for multi-step WTI futures price forecasting based on self-attention mechanism and spatial–temporal graph neural network," Resources Policy, Elsevier, vol. 85(PB).
    14. Ouyang, Zisheng & Lu, Min & Ouyang, Zhongzhe & Zhou, Xuewei & Wang, Ren, 2024. "A novel integrated method for improving the forecasting accuracy of crude oil: ESMD-CFastICA-BiLSTM-Attention," Energy Economics, Elsevier, vol. 138(C).
    15. Hongli Niu & Qiaoying Pan & Kunliang Xu, 2023. "Hybrid deep learning models with multi-classification investor sentiment to forecast the prices of China’s leading stocks," PLOS ONE, Public Library of Science, vol. 18(11), pages 1-24, November.
    16. Khan, Faridoon & Muhammadullah, Sara & Sharif, Arshian & Lee, Chien-Chiang, 2024. "The role of green energy stock market in forecasting China's crude oil market: An application of IIS approach and sparse regression models," Energy Economics, Elsevier, vol. 130(C).
    17. Butler, Sunil & Kokoszka, Piotr & Miao, Hong & Shang, Han Lin, 2021. "Neural network prediction of crude oil futures using B-splines," Energy Economics, Elsevier, vol. 94(C).
    18. Jean-Michel Sahut & Petr Hajek & Vladimir Olej & Lubica Hikkerova, 2025. "The role of news-based sentiment in forecasting crude oil price during the Covid-19 pandemic," Annals of Operations Research, Springer, vol. 345(2), pages 861-884, February.
    19. Liang, Qian & Lin, Qingyuan & Guo, Mengzhuo & Lu, Quanying & Zhang, Dayong, 2025. "Forecasting crude oil prices: A Gated Recurrent Unit-based nonlinear Granger Causality model," International Review of Financial Analysis, Elsevier, vol. 102(C).
    20. Omer Berat Sezer & Mehmet Ugur Gudelek & Ahmet Murat Ozbayoglu, 2019. "Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019," Papers 1911.13288, arXiv.org.

    More about this item

    NEP fields

    This paper has been announced in the following NEP Reports:

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:arx:papers:2508.06497. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: arXiv administrators (email available below). General contact details of provider: http://arxiv.org/ .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.