Carbon Trading Price Prediction of Three Carbon Trading Markets in China Based on a Hybrid Model Combining CEEMDAN, SE, ISSA, and MKELM

Carbon trading has been deemed as the most effective mechanism to mitigate carbon emissions. However, during market operation, competition among participants will inevitably occur; hence, precise forecasting of price (CTP) become a significant element in formulation strategies. This investigation established hybrid CTP framework combining complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), sample entropy (SE) method, improved salp swarm algorithm (ISSA), and multi-kernel extreme learning machine (MKELM) methods improve accuracy. Firstly, initial data sequence is disintegrated into several intrinsic functions (IMFs) residual by CEEMDAN method. Secondly, save calculation time, SE method utilized reconstruct IMFs new IMFs. Thirdly, are fed MKELM model, combing RBF poly kernel utilize their superior generalization abilities. The parameters model optimized ISSA, dynamic inertia weight chaotic local searching SSA enhance speed, convergence precision, well global ability. Guangdong, Shanghai, Hubei selected prove validity CEEMDAN-SE-ISSA-MKELM model. Through comparison analysis, performs best smallest MAPE RMSE values highest R2 value, which 0.76%, 0.53, 0.99, respectively, for Guangdong,. Thus, presented would be extensively applied future.