Research on Carbon Emission Allowance Price Prediction Based on Quadratic Decomposition

DOI: https://doi.org/10.62517/jse.202511503

Author(s)

Xueying Zhang*

Affiliation(s)

School of Mathematics and Statistics, North China University of Water Resources and Electric Power, Zhengzhou, Henan, China *Corresponding Author

Abstract

To address the pronounced non-stationarity and high complexity inherent in carbon emission allowance price series, this study introduces a hybrid forecasting framework grounded in secondary decomposition to enhance predictive precision. Initially, the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) algorithm is employed to disaggregate the original price data into multiple intrinsic mode functions (IMFs). These components are subsequently reconstructed into high-, medium-, and low-frequency clusters via K-means clustering based on fuzzy entropy. The high-frequency constituents undergo further decomposition using Variational Mode Decomposition (VMD) to mitigate volatility and noise. A Long Short-Term Memory (LSTM) network optimized by Particle Swarm Optimization (PSO) is applied to predict the refined high-frequency sub-sequences, while the medium- and low-frequency components are forecast using the XGBoost algorithm for its efficiency in handling structured sequences. The final prediction is derived through the aggregation of all reconstructed subsequences. Empirical validation using carbon market data from Hubei demonstrates that the proposed model achieves a reduction in mean absolute percentage error by up to 1.77 percentage points and improves the coefficient of determination by 15.31%, confirming its superior accuracy and robustness against benchmark models.

Keywords

Secondary Decomposition; Fuzzy Entropy; Carbon Price Forecasting; LSTM Model; Multi-Frequency Combination

References

[1]Li M Y, Wang J, Tian Y. Research on the development of China's economic industries under the carbon peak and carbon neutrality goals. Global Energy Interconnection, 2024, 7(06): 629-639. [2]Wang K, Lü C. Global and Chinese carbon market review and outlook (2025). Journal of Beijing Institute of Technology (Social Sciences Edition), 2025, 27(02): 19-36. [3]Huang Y X. Research on the volatility of China's carbon price based on the Skew-t-EGARCH model. China Forestry Economics, 2022, 173(02): 92-96. [4]Zhou L. Research on the price forecasting of the national carbon emission trading market based on a combined forecasting model. Jinan: Shandong University, 2023. [5]Huang J K. The impact of economic policy uncertainty on the volatility of Shenzhen's carbon emission trading market: An analysis based on the GARCH-MIDAS model. China Journal of Commerce, 2024, (12): 121-125. [6]Wen L, Liu Y. A research about Beijing's carbon emissions based on the IPSO-BP model. Environmental Progress & Sustainable Energy, 2017, 36(2): 428-434. [7]Huang X, He G Y, Sun Y X. Research on carbon trading price forecasting based on RF-PSO-LSTM model. Journal of Hunan City University (Natural Science Edition), 2024, 33(06): 61-66. [8]Wang Y F, Wang N, Zhang S H. Hybrid forecasting of EU carbon price based on multi-source data features and multi-scale analysis. Journal of Systems Science and Mathematical Sciences, 2024: 1-24. [9]Ma Z L. Carbon price forecasting based on decomposition and error correction with TCN-LSTM-Attention model. Shanxi University, 2024. [10]Chen Y, Zhang M, Gao J. Carbon emission allowance trading price prediction based on CEEMDAN-SE-GRU. Information System Engineering, 2024, (11): 123-126. [11]Duan J T, Yang X Z. Research on carbon price prediction method integrating deep learning and signal decomposition. The World of Survey and Research, 2024.