A hybrid convolutional neural network–bidirectional long short-term memory–attention framework for carbon price forecasting with residual error correction
Carbon trading markets are important mechanisms for reducing greenhouse gas emissions; however, carbon prices are highly volatile due to the influence of energy markets, economic conditions, regulatory policies, and investor sentiment. This study proposes a hybrid machine learning and natural language processing (ML–NLP) framework for interval-based carbon price forecasting by integrating structured market indicators with sentiment features extracted from financial news, policy announcements, and market discussions. The proposed model combines convolutional neural networks, bidirectional long short-term memory networks, and an attention mechanism to capture complex temporal dependencies and feature interactions, along with a residual error correction module to improve forecasting accuracy. The framework is evaluated using real-world European Union and China Emission Trading Systems carbon allowance price datasets. The results demonstrate that the proposed approach effectively captures nonlinear carbon market patterns and improves prediction performance, reducing mean absolute error from 2.84 €/tCO2 to 1.76 €/tCO2 after residual correction. The findings confirm that integrating deep learning, sentiment analysis, and residual learning provides an effective approach for reliable carbon price forecasting.
Conteh, F., Takahashi, H., Hemeida, A. M., Krishnan, N., Mikhaylov, A., & Senjyu, T. (2021). Analysis of hybrid grid-connected renewable power generation for sustainable electricity supply in Sierra Leone. Sustainability, 13(20), 11435. https://doi.org/10.3390/su132011435
Dong, H., Hu, Y., Yang, Y., & Jiang, W. (2023). A multi-strategy integration prediction model for carbon price. Energies, 16(12), 4613. https://doi.org/10.3390/en16124613
Feng, M., Duan, Y., Wang, X., Zhang, J., & Ma, L. (2023). Carbon price prediction based on decomposition technique and extreme gradient boosting optimized by the grey wolf optimizer algorithm. Scientific Reports, 13(1), 18447. https://doi.org/10.1038/s41598-023-45524-2
Gaber, E. S., Mikhaylov, A., Baranyai, N., Ahmed, M., & Hemeida, M. (2025). Optimal load frequency control system for two-area connected via AC/DC link using cuckoo search algorithm. Virtual Reality & Intelligent Hardware, 7(3), 299-316. https://doi.org/10.1016/j.vrih.2025.03.006
Huang, Y., Yona, A., Takahashi, H., Hemeida, A. M., Mandal, P., Mikhaylov, A., Senjyu, T., & Lotfy, M. E. (2021). Energy management system optimization of drug store electric vehicles charging station operation. Sustainability, 13(11), 6163. https://doi.org/10.3390/su13116163
Irshad, A. S., Zakir, M. N., Rashad, S. S., Lotfy, M. E., Mikhaylov, A., Elkholy, M. H., Pinter, G., & Senjyu, T. (2024). Comparative analyses and optimizations of hybrid biomass and solar energy systems based upon a variety of biomass technologies. Energy Conversion and Management: X, 23, 100640. https://doi.org/10.1016/j.ecmx.2024.100640
Mao, S., & Zeng, X. J. (2023). SimVGNets: similarity-based visibility graph networks for carbon price forecasting. Expert Systems with Applications, 230, 120647. https://doi.org/10.1016/j.eswa.2023.120647
Mujiono, E. P. U., Mukhlash, I., Pradana, Y. A., & Putri, E. R. M. (2025). Carbon price prediction by incorporating fossil fuel prices using LSTM with temporal pattern attention. Science and Technology Indonesia, 10(3), 856-865. https://doi.org/10.26554/sti.2025.10.3.856-865
Qin, C., Qin, D., Jiang, Q., & Zhu, B. (2024). Forecasting carbon price with attention mechanism and bidirectional long short-term memory network. Energy, 299, 131410. https://doi.org/10.1016/j.energy.2024.131410
Sayed, G. I., Abd El-Latif, E. I., Darwish, A., Snasel, V., & Hassanien, A. E. (2024). An optimized and interpretable carbon price prediction: Explainable deep learning model. Chaos, Solitons & Fractals, 188, 115533. https://doi.org/10.1016/j.chaos.2024.115533
Sediqi, M. M., Nakadomari, A., Mikhaylov, A., Krishnan, N., Lotfy, M. E., Yona, A., & Senjyu, T. (2022). Impact of time-of-use demand response program on optimal operation of Afghanistan real power system. Energies, 15(1), 296. https://doi.org/10.3390/en15010296
Shaikh, Z. A., Kraikin, A., Mikhaylov, A., & Pinter, G. (2022). Forecasting stock prices of companies producing solar panels using machine learning methods. Complexity, 2022(1), 9186265. https://doi.org/10.1155/2022/9186265
Shi, H., Wei, A., Xu, X., Zhu, Y., Hu, H., & Tang, S. (2024). A CNN-LSTM based deep learning model with high accuracy and robustness for carbon price forecasting: A case of Shenzhen's carbon market in China. Journal of Environmental Management, 352, 120131. https://doi.org/10.1016/j.jenvman.2024.120131
Sun, K., & Fan, X. (2026). Modeling Carbon Emission Allowance Prices Using Multi-Scale Decomposition and Integrated Deep Learning Approaches. Frontiers in Business, Economics and Management, 22(1), 113-119. https://doi.org/10.54097/enkwnq29
Sun, W., & Zhang, J. (2022). A novel carbon price prediction model based on optimized least square support vector machine combining characteristic-scale decomposition and phase space reconstruction. Energy, 253, 124167. https://doi.org/10.1016/j.energy.2022.124167
Wang, M. C., Chang, T., Mikhaylov, A., & Linyu, J. (2024). A measure of quantile-on-quantile connectedness for the US treasury yield curve spread, the US Dollar, and gold price. The North American Journal of Economics and Finance, 74, 102232. https://doi.org/10.1016/j.najef.2024.102232
Wang, Y., Qin, L., Wang, Q., Chen, Y., Yang, Q., Xing, L., & Ba, S. (2023). A novel deep learning carbon price short-term prediction model with dual-stage attention mechanism. Applied Energy, 347, 121380. https://doi.org/10.1016/j.apenergy.2023.121380
Wang, Y., Wang, Z., Kang, X., & Luo, Y. (2023). A novel interpretable model ensemble multivariate fast iterative filtering and temporal fusion transform for carbon price forecasting. Energy Science & Engineering, 11(3), 1148-1179. https://doi.org/10.1002/ese3.1380
Wu, H., & Du, P. (2024). Dual-stream transformer-attention fusion network for short-term carbon price prediction. Energy, 311, 133374. https://doi.org/10.1016/j.energy.2024.133374
Xu, W., Wang, J., Zhang, Y., Li, J., & Wei, L. (2025). An optimized decomposition integration framework for carbon price prediction based on multi-factor two-stage feature dimension reduction. Annals of Operations Research, 345(2), 1229-1266. https://doi.org/10.1007/s10479-022-04858-2
Yang, P., Wang, Y., Zhao, S., Chen, Z., & Li, Y. (2023). A carbon price hybrid forecasting model based on data multi-scale decomposition and machine learning. Environmental Science and Pollution Research, 30(2), 3252-3269. https://doi.org/10.1007/s11356-022-22286-4
Yang, Y., Guo, H., Jin, Y., & Song, A. (2021). An ensemble prediction system based on artificial neural networks and deep learning methods for deterministic and probabilistic carbon price forecasting. Frontiers in Environmental Science, 9, 740093. https://doi.org/10.3389/fenvs.2021.740093
Yun, P., Huang, X., Wu, Y., & Yang, X. (2023). Forecasting carbon dioxide emission price using a novel mode decomposition machine learning hybrid model of CEEMDAN-LSTM. Energy Science & Engineering, 11(1), 79-96. https://doi.org/10.1002/ese3.1304
Zhang, K., Yang, X., Wang, T., Thé, J., Tan, Z., & Yu, H. (2023). Multi-step carbon price forecasting using a hybrid model based on multivariate decomposition strategy and deep learning algorithms. Journal of Cleaner Production, 405, 136959. https://doi.org/10.1016/j.jclepro.2023.136959
Zhang, H., Zhou, L., & Chen, Y. (2025). Carbon price fluctuation prediction using a hybrid statistics and machine learning framework. Energy, 324, 135581. https://doi.org/10.1016/j.energy.2025.135581
Zhang, T., & Tang, Z. (2023). Multi-step carbon price forecasting based on a new quadratic decomposition ensemble learning approach. Frontiers in Energy Research, 10, 991570. https://doi.org/10.3389/fenrg.2022.991570
Zhang, X., Yang, K., Lu, Q., Wu, J., Yu, L., & Lin, Y. (2023). Predicting carbon futures prices based on a new hybrid machine learning: Comparative study of carbon prices in different periods. Journal of Environmental Management, 346, 118962. https://doi.org/10.1016/j.jenvman.2023.118962
Zou, S., & Zhang, J. (2024). A carbon price ensemble prediction model based on secondary decomposition strategies and bidirectional long short-term memory neural network by an improved particle swarm optimization. Energy Science & Engineering, 12(6), 2568-2590. https://doi.org/10.1002/ese3.1769