Machine Learning for New Energy Vehicle Sales Prediction and Carbon Emission Impact

Abstract

China's new energy vehicle market is in a booming stage, with the number of new energy vehicle buyers increasing year by year. The promotion of new energy vehicles is of great significance in reducing carbon emissions, environmental protection and promoting the construction of a clean and low-carbon society. In-depth analysis of market demand through the use of data mining and machine learning algorithms can help companies improve the market competitiveness and promotion of new energy vehicles. At the same time, in-depth study of the reduction effect of new energy vehicles on carbon emissions will help to comprehensively assess their environmental benefits and provide a scientific basis for the government to formulate carbon emission reduction policies and promote the development of the new energy vehicle industry. In this paper, we screened the multifaceted factors affecting the sales of new energy vehicles by using the random forest model, and then used the support vector regression machine to predict the sales of different brands of new energy vehicles in the next two months, and further verified the accuracy and reliability of the model after the effect evaluation on the test set. In addition, this paper establishes a panel regression model of the carbon emission reduction effect of new energy vehicles by introducing control variables and using panel data from 31 provinces, autonomous regions and municipalities in China from 2017 to 2022. The panel regression model verifies that the development of the new energy vehicle industry is conducive to promoting carbon emission reduction from the overall national level.