Forecasting Commodity Prices Using Deep Learning Techniques: An Empirical Evidence from India

Year : 2025 | Volume : 16 | Issue : 03 | Page : 08 12
    By

    Amrinder Kaur,

  • Archana Goel,

  1. Research Scholar, Department of Marketing and Finance Programme, Chitkara Business School, Chitkara University, Punjab, India
  2. Associate Professor, Department of Marketing and Finance Programme, Chitkara Business School, Chitkara University, Punjab, India

Abstract

Commodity price forecasting is instrumental in financial markets, providing framework for investment choices and risk management practices. Traditional models, including statistical and machine learning approaches, have limitations in capturing the nonlinear and volatile nature of commodity prices. Deep learning (DL) techniques have emerged as promising alternatives, leveraging advanced neural networks to enhance predictive accuracy. This study presents a thorough and comprehensive examination of deep learning applications in commodity price prediction, focusing on the Indian market. It highlights existing research gaps such as insufficient exploration of hyperparameter optimization, lack of risk-averse strategies, and limited integration of macroeconomic indicators and financial news sentiment analysis. Furthermore, most studies rely on short-term datasets, overlooking seasonal and annual trends that significantly influence commodity prices. By examining various deep learning models, including LSTM, Convolutional Neural Networks (CNN), and Reinforcement Learning (RL), this study aims to identify the most effective models for forecasting Indian commodity prices. Additionally, it compares deep learning methods with traditional forecasting techniques such as ARIMA and GARCH. The findings suggest that deep learning models, particularly that incorporating sentiment analysis from financial news, exceed the performance of conventional models in forecasting precision and economic efficiency. The study also emphasizes the need for interval forecasting to provide more comprehensive insights for investors and policymakers. Future research should focus on refining model architectures, incorporating broader datasets, and integrating hybrid models to optimize predictive performance.

Keywords: Commodity price forecasting, deep learning (DL), Indian commodity market, machine learning (ML), financial markets, time series analysis, sentiment analysis

[This article belongs to Journal of Computer Technology & Applications ]

How to cite this article:
Amrinder Kaur, Archana Goel. Forecasting Commodity Prices Using Deep Learning Techniques: An Empirical Evidence from India. Journal of Computer Technology & Applications. 2025; 16(03):08-12.
How to cite this URL:
Amrinder Kaur, Archana Goel. Forecasting Commodity Prices Using Deep Learning Techniques: An Empirical Evidence from India. Journal of Computer Technology & Applications. 2025; 16(03):08-12. Available from: https://journals.stmjournals.com/jocta/article=2025/view=226934


References

  1. Moreno C, Saavedra C, Ulloa B. Commodity price cycles and financial stability. Available at SSRN 2594266. 2014 Aug.
  2. Aldabagh H, Zheng X, Mukkamala R. A hybrid deep learning approach for crude oil price prediction. J Risk Financ Manag. 2023 Dec 6; 16(12): 503.
  3. Zhang JL, Zhang YJ, Li DZ, Tan ZF, Ji JF. Forecasting day-ahead electricity prices using a new integrated model. Int J Electr Power Energy Syst. 2019 Feb 1; 105: 541–8.
  4. Cortez CT, Saydam S, Coulton J, Sammut C. Alternative techniques for forecasting mineral commodity prices. Int J Min Sci Technol. 2018 Mar 1; 28(2): 309–22.
  5. Ojugo AA, Yoro RE. Predicting Futures price and contract portfolios using the ARIMA model: a case of Nigeria’s Bonny light and forcados. Quant Econ Manag Stud. 2020 Aug 22; 1(4): 237–48.
  6. Kamdem JS, Essomba RB, Berinyuy JN. Deep learning models for forecasting and analyzing the implications of COVID-19 spread on some commodities markets volatilities. Chaos Solit Fractals. 2020 Nov 1; 140: 110215.
  7. Gupta V, Pandey A. Crude oil price prediction using LSTM networks. International Journal of Computer and Information Engineering. 2018; 12(3): 226–30.
  8. Cen Z, Wang J. Crude oil price prediction model with long short term memory deep learning based on prior knowledge data transfer. Energy. 2019 Feb 15; 169: 160–71.
  9. Zhang H, Nguyen H, Vu DA, Bui XN, Pradhan B. Forecasting monthly copper price: A comparative study of various machine learning-based methods. Resour Policy. 2021 Oct 1; 73: 102189.
  10. Khurana V, Gahalawat M, Kumar P, Roy PP, Dogra DP, Scheme E, Soleymani M. A survey on neuromarketing using EEG signals. IEEE Trans Cogn Dev Syst. 2021 Mar 12; 13(4): 732–49.
  11. Mariono M, Syaharuddin S, Ashraf S, Fadugba SE. Analyzing Social Media Sentiment Toward Specific Commodities for Forecasting Price Movements in Commodity Markets. BAREKENG: Jurnal Ilmu Matematika dan Terapan. 2025 Jan 13; 19(1): 199–214.
  12. Chen X, Li B, Wang J, Zhao Y, Xiong Y. Integrating EMD with multivariate LSTM for time series QoS prediction. In 2020 IEEE International Conference on Web Services (ICWS). 2020 Oct 19; 58–65.
  13. Pei Y, Huang CJ, Shen Y, Wang M. A novel model for spot price forecast of natural gas based on temporal convolutional network. Energies. 2023 Feb 28; 16(5): 2321. 14. Han L, Li Z, Yin L. The effects of investor attention on commodity futures markets. J Futures Mark. 2017 Oct; 37(10): 1031–49.
Regular Issue Subscription Review Article
Volume 16
Issue 03
Received 28/04/2025
Accepted 10/07/2025
Published 07/08/2025
Publication Time 101 Days
56


My IP

PlumX Metrics