基于变权组合模型的碳排放量预测

Carbon Emission Prediction Based on Variable Weight Combination Model

碳排放量的预测一直是国内外人们关注的热点,为了进一步提高碳排放量预测模型的准确性,考虑多种因素对碳排放量的影响,利用支持向量机回归、岭回归和BP神经网络三种传统单项碳排放量预测模型,结合误差倒数法构建了一种变权组合模型,并利用新模型预测我国2022—2026年的碳排放量。实证结果显示,组合模型的拟合精度和预测精度分别为99.26%和99.34%,组合模型对比3种单项模型有更高的精度。组合模型的预测结果显示,到2026年,我国碳排放增速较现在有所放缓,以1.8%的速度保持增长。

The prediction of carbon emissions has always been a hot spot of people's attention at home and abroad.In order to further improve the accuracy of the carbon emission prediction model,considering the impact of multiple factors on carbon emissions,this paper uses three traditional single-item carbon emission prediction models of Support Vector Regression,Ridge Regression and BP Neural Network and combines with the inverse of the error method to construct a variable weight combination model,and uses the new model to predict China's carbon emissions from 2022 to 2026.The empirical results show that the fitting and prediction accuracy of the combination model is 99.26%and 99.34%,respectively,and the combination model has higher accuracy than the three single models.The prediction results of the combination model show that by 2026,the growth rate of China's carbon emissions has slowed down compared with the present,and maintains the growth rate at 1.8%.