A novel hierarchical carbon price forecasting model with local and overall perspectives

Existing carbon price decomposition methods make effective predictions, promote energy saving and emission reduction, and play an increasingly important role in carbon trading platforms. However, few studies have been conducted on the reorganization methods and different perspective treatments of the decomposition components. In this paper, a new component fusion method is introduced, based on this, a hierarchical carbon price prediction model with two levels—one for a local perspective and one for an overall one—is developed. Firstly, the carbon price data are decomposed and the resulting components are subjected to deviation sample entropy fusion, which classifies them into high, medium, and low frequencies according to the physical significance of the entropy values. Next, fine-grained predictions are performed for the high, medium and low frequency components, defining this step as the local layer. Subsequently, the decomposition error correction is proposed to obtained the data of the overall layer, and a secondary decomposition of this data is done. Finally, the prediction values of the two levels are summed to obtain the final prediction results. The experimental results in three markets, Guangdong, Tianjin and Beijing, show that the proposed fusion method can directly find the best component recombination scheme and the model prediction ability is better than the conventional secondary decomposition model.