A Patch-Based Transformer Approach to Nonlinear Dynamics Natural Gas Price Forecasting
DOI:
https://doi.org/10.12962/j24775401.ijcsam.v11i2.8849Keywords:
Natural Gas, Time Series Forecasting, Transformer Model, Energy Market PredictionAbstract
Natural gas prices are a critical economic indicator influencing various sectors of the global economy. Accurate forecasting is essential for effective policy formulation and strategic decision making. However, natural gas price movements often exhibit complex non-linear patterns that traditional statistical time series models fail to capture. Furthermore, many deep learning architectures struggle to effectively model these intricate dynamics. To address this challenge, we employ the Patch-Based Transformer (PatchTST) model for natural gas price forecasting. The comparative results reveal that PatchTST achieves substantially higher predictive accuracy than both statistical and other deep learning models. Its Transformer-based architecture, combined with patching and channel independence, enables the model to effectively capture both temporal dependencies and localized variations. The model achieved mean squared error (MSE) and mean absolute percentage error (MAPE) values of 0.1176 and 7.57%, respectively. These findings demonstrate that PatchTST provides robust and precise forecasts, offering valuable insights for decision-making in the energy market
References
[1] United Nations, “Paris agreement,” 2015, available online.
[2] International Energy Agency, World Energy Outlook 2019. OECD, 2019.
[3] S. Paltsev, “Projecting energy and climate for the 21st century,” Economics of Energy & Environmental Policy, vol. 9, no. 1, Jan. 2020.
[4] BP, “Bp energy outlook 2019,” 2019.
[5] M. Boeck and T. O. Z¨orner, “Natural gas prices, inflation expectations, and the pass-through to euro area inflation,” Energy Economics, vol. 141, p. 108061, Jan. 2025.
[6] W. Cui, D. Wu, Q. Huang, and S. Yang, “The dynamic evolution mechanism of liquefied natural gas trade dependency networks: International implications of the russia-ukraine conflict,” Sustainable Futures, vol. 9, p. 100730, Jun. 2025.
[7] G. E. P. Box and G. M. Jenkins, Time Series Analysis: Forecasting and Control. San Francisco: Holden-Day, 1970.
[8] J. D. Hamilton, “A new approach to the economic analysis of nonstationary time series and the business cycle,” Econometrica, pp. 357–384.
[9] S. Gao, C. Hou, and B. H. Nguyen, “Forecasting natural gas prices using highly flexible time-varying parameter models,” Economic Modelling, vol. 105, p. 105652, Dec. 2021.
[10] J. L. Elman, “Finding structure in time,” Cognitive Science, vol. 14, no. 2, pp. 179–211.
[11] S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780.
[12] Y. Lin et al., “Forecasting natural gas prices using a novel hybrid model: Comparative study of different sliding windows,” Energy, vol. 329, p. 136607, Aug. 2025.
[13] A. Vaswani et al., “Attention is all you need,” in Advances in Neural Information Processing Systems, vol. 30.
[14] Y. Nie, N. H. Nguyen, P. Sinthong, and J. Kalagnanam, “A time series is worth 64 words: Long-term forecasting with transformers,” Mar. 2023.
[15] Z. Wu and N. E. Huang, “Ensemble empirical mode decomposition: a noise-assisted data analysis method,” Advances in Adaptive Data Analysis, vol. 1, no. 1, pp. 1–41.
[16] S. Bai, J. Z. Kolter, and V. Koltun, “An empirical evaluation of generic convolutional and recurrent networks for sequence modeling,” Apr. 2018.
[17] S. Woo, J. Park, J.-Y. Lee, and I. S. Kweon, “Cbam: Convolutional block attention module,” pp. 3–19.
[18] Y. Zheng, J. Luo, J. Chen, Z. Chen, and P. Shang, “Natural gas spot price prediction research under the background of russia-ukraine conflict — based on fs-ga-svr hybrid model,” Journal of Environmental Management, vol. 344, p. 118446, Oct. 2023.
[19] V. N. Vapnik, Statistical Learning Theory. Wiley, 1998.
[20] H. Wu, J. Xu, J. Wang, and M. Long, “Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting,” in Advances in Neural Information Processing Systems, vol. 34, pp. 22 419–22 430.
[21] H. Zhou et al., “Informer: Beyond efficient transformer for long sequence time-series forecasting,” in Proc. AAAI Conf. Artificial Intelligence, vol. 35, pp. 11 106–11 115.
[22] A. Zeng, M. Chen, L. Zhang, and Q. Xu, “Are transformers effective for time series forecasting?” Aug. 2022.
[23] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “Bert: Pre-training of deep bidirectional transformers for language understanding,” in Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). Association for Computational Linguistics, 2019, pp. 4171–4186. [Online]. Available: https://arxiv.org/abs/1810.04805
[24] A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly, J. Uszkoreit, and N. Houlsby, “An image is worth 16x16 words: Transformers for image recognition at scale,” arXiv preprint arXiv:2010.11929, 2020. [Online]. Available: https://arxiv.org/abs/2010.11929
[25] Investing.com, “Natural gas futures price today,” https://www.investing.com/commodities/natural-gas, [Accessed: 09-Oct-2025].
[26] A. M. De Livera, R. J. Hyndman, and R. D. Snyder, “Forecasting time series with complex seasonal patterns using exponential smoothing,” Journal of the American Statistical Association, vol. 106, no. 496, pp. 1513–1527, 2011.
[27] S. J. Taylor and B. Letham, “Forecasting at scale,” The American Statistician, vol. 72, no. 1, pp. 37–45, 2018.
[28] D. A. Dickey and W. A. Fuller, “Distribution of the estimators for autoregressive time series with a unit root,” Journal of the American statistical association, vol. 74, no. 366, pp. 427–431, 1979.
[29] H. Akaike, “Information theory and an extension of the maximum likelihood principle,” in Proceedings of the Second International Symposium on Information Theory. Budapest: Akademiai Kiado, 1973, pp. 267–281.
[30] G. E. Schwarz, “Estimating the dimension of a model,” The Annals of Statistics, vol. 6, no. 2, pp. 461–464, 1978.
[31] G. M. Ljung and G. E. P. Box, “On a measure of lack of fit in time series models,” Biometrika, vol. 65, no. 2, pp. 297–303, 1978.
[32] P. R. Winters, “Forecasting sales by exponentially weighted moving averages,” Management Science, vol. 6, no. 3, pp. 324–342, 1960.
[33] R. G. Brown, “Exponential smoothing for predicting demand,” Arthur D. Little Inc., Cambridge, MA, Tech. Rep., 1956.
[34] C. C. Holt, “Forecasting trends and seasonals by exponentially weighted moving averages,” Carnegie Institute of Technology, Pittsburgh, PA, Tech. Rep., 1957.
