Research On Carbon Emissions Permit Trading Price Prediction Considering Influencing Factors

In September 2020,China pledged to peak its carbon dioxide emissions by 2030 and strive for carbon neutrality by 2060.As an important policy tool to achieve the goal,carbon emission trading is a major institutional innovation to deal with climate change.Since 2013,China has established eight pilot carbon markets.In 2017,China announced the launch of the national carbon market construction,and on July 16,2021,the first carbon emission trading in the power generation industry has been launched offi cially.With the continuous development of carbon market,its financial attribute is increasingly prominent.As one of the core indicators of carbon financial market,carbon price not only affects the performance of emission reduction,but also affects the cost and future development of emission reduction subjects.Therefore,accurate prediction of carbon financial market can not only provides scientific decision-making basis for investors and regulatory authorities,but also effectively promotes the health y development of carbon financial market.On the basis of systematic review of existing studies,this paper analyzes and expounds related concepts such as carbon emission price and related theories such as carbon price prediction.Then,it makes theoreti cal analysis of relevant influencing factors of carbon emission right price.By constructing VAR model,conducting Granger causality test,impulse response analysis and variance decomposition,the dynamic characteristics of the influence of different variables on Beijing’s carbon price are analyzed and identified.In addition,the impact of other factors from large to small are EUA price,Shanghai-Shenzhen 300 price,oil price,coal price,natural gas price,Shanghai Composite Index,RMB/Euro exchange rate,RMB/US dollar exchange rate,SHIBOR.Based on the identification of the influencing factors of carbon emission price,a carbon emission price prediction model based on MEMD-SSALSSVM method is constructed,and the Beijing carbon market is taken as a typical representative to make prediction analysis.Firstly,the carbon emission price and influencing factors are decomposed by MEMD,and then the data is reconstructed.The original complex data is decomposed and reconstructed into stable and regular short-term,medium-term,long-term and trend components with different economic characteristics.On this basis,the least square support vector machine(LSSVM)model is used to predict each component,and the sparrow search algorithm(SSA)is used to optimize the prediction model.Finally,the final prediction result is obtained by integrating each prediction component.By comparing the prediction accuracy with other control models,it is found that the composite prediction model proposed in this paper has th e minimum prediction error,which indicates that the prediction model proposed in this paper has better prediction performance.Finally,the paper puts forward relevant policy suggestions to guide the reasonable formation of carbon emission trading price and promote the steady development of carbon emission trading market.