Analysis of the Factors Influencing Carbon Trading Prices in Shenzhen

Abstract

China's carbon emissions trading market is in the ascendant. As the country with the largest carbon emissions in the world, China is actively taking measures to deal with global climate change. China has put forward the long-term goal of reaching the peak of carbon emissions by 2030, achieving carbon neutralization by 2060 and shouldering important international social responsibilities. The construction and development of carbon emissions trading market is a key link in achieving the goal of carbon peak and carbon neutralization. The research on the influencing factors of China's regional carbon emissions trading price and the development of regional green finance can provide decision support for achieving the goal of carbon peak and carbon neutralization in China. Starting from the four latitudes of domestic and foreign economies, domestic and foreign energy prices, international carbon prices, and international major currencies and RMB exchange rate, this paper selects 13 independent variables to explore their impact on the dependent variable Shenzhen carbon trading average price. First of all, this paper uses lasso for regression modeling, and selects the λ that minimizes the mse into the model. Five features of the model are screened out. Among the remaining characteristics, the CSI 300 index has a greater positive effect on the Shenzhen carbon trading price, while the SSE 50 has a greater reverse effect on the Shenzhen carbon trading price. Secondly, in order to avoid using lasso to compress the model coefficients too small, this paper uses the elastic network model to fit the data, the fitting result is actually the ridge regression model, and the prediction mean square error of the test set is greater than that of lasso modeling. Finally, considering all the characteristics comprehensively, this paper uses the random forest algorithm to model the data, and the characteristics that make a great contribution to the carbon trading price in Shenzhen are EU emission quota and Guangdong carbon trading price. and the mean square error of the test set modeled by random forest algorithm is the smallest of the three models, and the prediction effect is the best.