%0 Journal Article
%T 全国碳市场碳交易价格的影响因素及预测研究<br>Research on Influencing Factors and Prediction of Carbon Trading Price in National Carbon Market
%A 郭建平
%A 邵烁
%J Modeling and Simulation
%P 304-313
%@ 2324-870X
%D 2024
%I Hans Publishing
%R 10.12677/MOS.2024.131029
%X 准确可靠的碳市场价格预测对于保证全国碳市场平稳运行和完善相关制度机制建设具有重要意义，本文提出了基于完全集合经验模态分解和频域解析分析组合方法的碳排放权交易价格预测模型。首先使用完全集合经验模态分解方法对全国碳市场价格序列施以分解，获取本征模态分量并进行重构，确定影响碳交易价格的主要影响因素；然后对重构后的高频分量、低频分量和趋势项分别进行快速傅里叶变换(FFT)，估计各分量的振幅、角频率和相位参数，构建全国碳市场交易价格的频域解析式并进行预测。实证结果表明，高频序列所代表的投机、供需不平衡以及天气状况等周期较短的因素对碳市场价格的波动影响不显著，低频序列以及趋势项所代表的碳市场内在交易机制和长期价值等因素对碳市场价格的影响占主导地位。与当前主流的预测方法相比，本文所提出的预测方法预测精度显著提升。<br />
Accurate and reliable carbon market price prediction is of great significance for ensuring the smooth operation of the national carbon market and improving the construction of relevant institu-tional mechanisms. In this paper, a carbon emission rights trading price prediction model based on the combination of complete set empirical mode decomposition and frequency domain analytical analysis is proposed. First, the complete set empirical mode decomposition method is used to de-compose the national carbon market price series, obtain the eigenmodal components, reconstruct them, and determine the main influencing factors of carbon trading prices. Then, the reconstructed high-frequency component, low-frequency component and trend term are respectively evaluated by fast Fourier transform (FFT), and the amplitude, angular frequency and phase parameters of each component are estimated. The frequency domain analytic formula of the national carbon market trading price is constructed and forecasted. The empirical results show that short-cycle factors such as speculation, supply-demand imbalance and weather conditions represented by high-frequency series have no significant impact on the volatility of carbon market prices, while factors such as in-ternal trading mechanism and long-term value of carbon market represented by low-frequency se-ries and trend items dominate the impact on carbon market prices. Compared with the current mainstream forecasting methods, the prediction accuracy of the proposed forecasting method is significantly improved.
%K 全国碳市场，快速傅立叶变换，完全集合经验模态分解，本征模态分量，频域解析<br>National Carbon Market
%K Fast Fourier Transform
%K Complete Ensemble Empirical Mode 
Decomposition
%K Intrinsic Mode Component
%K Frequency Domain Resolution
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=79166