融合多源信息的碳价时滞组合预测

doi:10.3778/j.issn.1002-8331.2401-0109

摘要/Abstract

摘要： 碳价是碳市场的核心要素，碳价波动受到众多因素及其时滞效应的影响。为精准预测全国碳市场碳排放配额（Chinese emission allowances，CEA）价格，从关联碳市场、经济发展、国外能源、国内能源和人民币汇率五个维度选取结构化影响因素，从经济政策、环境影响和用户意愿三个维度爬取来自百度搜索引擎的非结构化影响因素，然后引入MIV-BP模型筛选主要的影响因素，并基于最大信息系数（maximum information coefficient，MIC）对碳价以及多源影响因素进行时滞估计。在此基础上，构建融合多源信息的碳价时滞组合预测模型MIC-LSTM-BP，并和基准模型LSTM、BP、LSTM-BP以及时滞基准模型MIC-LSTM、MIC-BP、MIC-LSTM-GBDT进行对比分析，以验证新模型的有效性。结果表明，时滞信息的引入有助于提升模型的预测精度；相较于基准模型和时滞基准模型，MIC-LSTM-BP模型预测CEA价格精度最高，价格波动追随能力最好。

关键词: 全国碳市场, 多源信息, 影响因素筛选, 时滞估计, 组合预测, MIC-LSTM-BP模型

Abstract: The carbon price is the core element of the carbon market, and its fluctuations are influenced by numerous factors and their time-delay effects. To precisely forecast the price of Chinese emission allowances (CEA) in the national carbon market, structured influencing factors are selected from five dimensions: related carbon markets, economic development, foreign energy, domestic energy, and the RMB exchange rate. Unstructured influencing factors are crawled from three dimensions: economic policy, environmental impact, and user preference, using the Baidu search engine. The MIV-BP model is introduced to screen the main influencing factors and carbon price, and the time-delay of multi-source factors are estimated based on maximum information coefficient (MIC). On this basis, a time-delay combined prediction model for carbon price is constructed based on MIC, LSTM and BP. Compared with baseline models such as LSTM, BP, LSTM-BP, and time-delay baseline models MIC-LSTM, MIC-BP, and MIC-LSTM-GBDT, the effectiveness of the new model is verified. The results indicate that the introduction of time-delay information is helpful to improve the prediction accuracy of the model. Compared with the baseline and time-delay baseline models, the MIC-LSTM-BP model has the highest precision in forecasting CEA prices and the best capability to follow price volatility.

Key words: national carbon market, multi-source information, influence factor screening, time-delay estimation, combined prediction, MIC-LSTM-BP model

邹艳, 王淑平, 李欣岷, 龚科. 融合多源信息的碳价时滞组合预测[J]. 计算机工程与应用, 2025, 61(10): 350-360.

ZOU Yan, WANG Shuping, LI Xinmin, GONG Ke. Carbon Price Forecasting Based on Multi-Source Information Fusion and Time-Delay Effect[J]. Computer Engineering and Applications, 2025, 61(10): 350-360.

参考文献

[1] YIN Y K, JIANG Z H, LIU Y Z, et al. Factors affecting carbon emission trading price: evidence from China[J]. Emerging Markets Finance and Trade, 2019, 55(15): 3433-3451.
[2] 周坤, 高晓辉, 李廉水. 基于EMD-XGB-ELM和FSGM双元处理的碳排放交易价格集成预测[J]. 中国管理科学, 2024, 32(10): 325-334.
ZHOU K, GAO X H, LI L S. Integrated carbon emission trading price prediction based on EMD-XGB-ELM and FSGM from the perspective of dual processing[J]. Chinese Journal of Management Science, 2024, 32(10): 325-334.
[3] 秦全德, 黄兆荣, 黄凯珊. 一种基于局部回归的多尺度碳市场价格预测模型研究[J]. 运筹与管理, 2022, 31(1): 107-114.
QIN Q D, HUANG Z R, HUANG K S. A multi-scale carbon price forecasting model with local regression approach[J]. Operations Research and Management Science, 2022, 31(1): 107-114.
[4] 范丽伟, 董欢欢, 渐令. 基于滚动时间窗的碳市场价格分解集成预测研究[J]. 中国管理科学, 2023, 31(1): 277-286.
FAN L W, DONG H H, JIAN L. A decomposition ensemble model with sliding time window for forecasting carbon market prices[J]. Chinese Journal of Management Science, 2023, 31(1): 277-286.
[5] 张大斌, 张博婷, 凌立文, 等. 基于二次分解聚合策略的我国碳交易价格预测[J]. 系统科学与数学, 2022, 42(11): 3094-3106.
ZHANG D B, ZHANG B T, LING L W, et al. Carbon price forecasting based on secondary decomposition and aggregation strategy[J]. Journal of Systems Science and Mathematical Sciences, 2022, 42(11): 3094-3106.
[6] LIN B Q, XU B. A non-parametric analysis of the driving factors of China’s carbon prices[J]. Energy Economics, 2021, 104: 105684.
[7] 魏宇, 张佳豪, 陈晓丹. 基于DMS和DMA的我国碳排放权交易价格预测方法: 来自湖北碳市场的经验证据[J]. 系统工程, 2022, 40(4): 1-16.
WEI Y, ZHANG J H, CHEN X D. Forecasting China’s carbon trading price in a DMS and DMA framework: evidence from the Hubei carbon market[J]. Systems Engineering, 2022, 40(4): 1-16.
[8] 蒋锋, 彭紫君. 基于混沌PSO优化BP神经网络的碳价预测[J]. 统计与信息论坛, 2018, 33(5): 93-98.
JIANG F, PENG Z J. Forecasting of carbon price based on BP neural network optimized by chaotic PSO algorithm[J]. Journal of Statistics and Information, 2018, 33(5): 93-98.
[9] 崔焕影, 窦祥胜. 基于EMD-GA-BP与EMD-PSO-LSSVM的中国碳市场价格预测[J]. 运筹与管理, 2018, 27(7): 133-143.
CUI H Y, DOU X S. Carbon price forecasts in Chinese carbon trading market based on EMD-GA-BP and EMD-PSO-LSSVM[J]. Operations Research and Management Science, 2018, 27(7): 133-143.
[10] NGUYEN T H, SHIRAI K, VELCIN J. Sentiment analysis on social media for stock movement prediction[J]. Expert Systems with Applications, 2015, 42(24): 9603-9611.
[11] 刘金培, 郭艺, 陈华友, 等. 基于非结构数据流行学习的碳价格多尺度组合预测[J]. 控制与决策, 2019, 34(2): 279-286.
LIU J P, GUO Y, CHEN H Y, et al. Multi-scale combined forecast of carbon price based on manifold learning of unstructured data[J]. Control and Decision, 2019, 34(2): 279-286.
[12] 刘金培, 张了丹, 朱家明, 等. 非结构性数据驱动的混合分解集成碳交易价格组合预测[J]. 运筹与管理, 2023, 32(3): 149-154.
LIU J P, ZHANG L D, ZHU J M, et al. Unstructured data driven carbon price combined forecast based on hybrid decomposition-integration[J]. Operations Research and Management Science, 2023, 32(3): 149-154.
[13] PAN D, ZHANG C, ZHU D D, et al. Carbon price forecasting based on news text mining considering investor attention[J]. Environmental Science and Pollution Research International, 2023, 30(11): 28704-28717.
[14] LIU M Y, YING Q W. The role of online news sentiment in carbon price prediction of China’s carbon markets[J]. Environmental Science and Pollution Research, 2023, 30(14): 41379-41387.
[15] WANG P, LIU J P, TAO Z F, et al. A novel carbon price combination forecasting approach based on multi-source information fusion and hybrid multi-scale decomposition[J]. Engineering Applications of Artificial Intelligence, 2022, 114: 105172.
[16] LI J M, LIU D H. Carbon price forecasting based on secondary decomposition and feature screening[J]. Energy, 2023, 278: 127783.
[17] ZHANG X, ZHANG C C, WEI Z Q. Carbon price forecasting based on multi-resolution singular value decomposition and extreme learning machine optimized by the moth-flame optimization algorithm considering energy and economic factors[J]. Energies, 2019, 12(22): 4283.
[18] SUN W, WANG Y W. Factor analysis and carbon price prediction based on empirical mode decomposition and least squares support vector machine optimized by improved particle swarm optimization[J]. Carbon Management, 2020, 11(3): 315-329.
[19] ZHAO L T, MIAO J, QU S, et al. A multi-factor integrated model for carbon price forecasting: market interaction promoting carbon emission reduction[J]. Science of the Total Environment, 2021, 796: 149110.
[20] HUANG Y C, HE Z. Carbon price forecasting with optimization prediction method based on unstructured combination[J]. Science of the Total Environment, 2020, 725: 138350.
[21] CAO Y, ZHA D L, WANG Q W, et al. Probabilistic carbon price prediction with quantile temporal convolutional network considering uncertain factors[J]. Journal of Environmental Management, 2023, 342: 118137.
[22] LI D, REN X F. Carbon price prediction based on LsOALEO feature selection and time-delay least angle regression[J]. Journal of Cleaner Production, 2023, 416: 137853.
[23] 吕靖烨, 王腾飞. 我国碳排放权市场价格波动的长期记忆性和杠杆效应研究: 以湖北碳排放权交易中心为例[J]. 价格月刊, 2019(10): 29-36.
LYU J Y, WANG T F. Study on long-term memory and leverage effect of price fluctuations in China’s carbon emission right market: taking Hubei Carbon Emission Right Trading Center as an example[J]. Prices Monthly, 2019(10): 29-36.
[24] CUI S Z, WANG D J, YIN Y Q, et al. Carbon trading price prediction based on a two-stage heterogeneous ensemble method[J]. Annals of Operations Research, 2025, 345: 953-977.
[25] XU W J, WANG J J, ZHANG Y, et al. An optimized decomposition integration framework for carbon price prediction based on multi-factor two-stage feature dimension reduction[J]. Annals of Operations Research, 2025, 345: 1229-1266.
[26] 何芳, 王小川, 肖森予, 等. 基于MIV-BP型网络实验的房地产项目风险识别研究[J]. 运筹与管理, 2013, 22(2): 229-234.
HE F, WANG X C, XIAO S Y, et al. Research on risk recognition of real estate projects based on MIV-BP neural network test[J]. Operations Research and Management Science, 2013, 22(2): 229-234.
[27] 王小燕, 周思敏, 徐晓莉, 等. 基于图结构自适应Lasso的碳排放权价格影响因素分析[J]. 统计与信息论坛, 2022, 37(4): 73-83.
WANG X Y, ZHOU S M, XU X L, et al. Influencing factors analysis for the price of China’s regional carbon emissions based on graph-adaptive Lasso[J]. Journal of Statistics and Information, 2022, 37(4): 73-83.
[28] 郭文军. 中国区域碳排放权价格影响因素的研究: 基于自适应Lasso方法[J]. 中国人口·资源与环境, 2015, 25(S1): 305-310.
GUO W J. Factors impacting on the price of China’s regional carbon emissions based on adaptive Lasso method[J]. China Population, Resources and Environment, 2015, 25(S1): 305-310.
[29] 王倩, 高翠云. 中国试点碳市场间的溢出效应研究: 基于六元VAR-GARCH-BEKK模型与社会网络分析法[J]. 武汉大学学报 (哲学社会科学版), 2016, 69(6): 57-67.
WANG Q, GAO C Y. Research on the spillover effect of the pilot carbon trading markets in China: based on sextuple VAR-GARCH-BEKK model and social network analysis[J]. Wuhan University Journal (Philosophy & Social Sciences), 2016, 69(6): 57-67.
[30] 孙春. 中国碳市场与EU碳市场价格波动溢出效应研究[J]. 工业技术经济, 2018, 37(3): 97-105.
SUN C. Spillover effects of price fluctuation on China’s carbon market and EU carbon market[J]. Journal of Industrial Technological Economics, 2018, 37(3): 97-105.
[31] YUAN N N, YANG L. Asymmetric risk spillover between financial market uncertainty and the carbon market: a GAS-DCS-copula approach[J]. Journal of Cleaner Production, 2020, 259: 120750.
[32] WANG R, ZHAO X L, WU K R, et al. Examination of the transmission mechanism of energy prices influencing carbon prices: an analysis of mediating effects based on demand heterogeneity[J]. Environmental Science and Pollution Research, 2023, 30(21): 59567-59578.
[33] 王娜. 基于动态多元网络的中国碳价预测[J]. 统计研究, 2023, 40(1): 49-61.
WANG N. Carbon price prediction in China based on dynamic multivariate network[J]. Statistical Research, 2023, 40(1): 49-61.
[34] 易兰, 杨历, 李朝鹏, 等. 欧盟碳价影响因素研究及其对中国的启示[J]. 中国人口·资源与环境, 2017, 27(6): 42-48.
YI L, YANG L, LI Z P, et al. Impacts of multiple factors on EU carbon price and implications to China[J]. China Population, Resources and Environment, 2017, 27(6): 42-48.
[35] 杜子平, 刘富存. 我国区域碳排放权价格及其影响因素研究: 基于GA-BP-MIV模型的实证分析[J]. 价格理论与实践, 2018(6): 42-45.
DU Z P, LIU F C. Impacts on the price of regional carbon emissions based on GA-BP-MIV model[J]. Price: Theory & Practice, 2018(6): 42-45.
[36] 孙悦鹏, 孙延吉, 潘艳秋, 等. 基于BO-LSTM的低温甲醇洗净化气CO2含量预测[J]. 化工进展, 2025, 44(2): 688-697.
SUN Y P, SUN Y J, PAN Y Q, et al. Prediction of CO2 content in Rectisol purified gas based on BO-LSTM[J]. Chemical Industry and Engineering Progress, 2025, 44(2): 688-697.