Analysis and Prediction of Influencing Factors of Carbon Emission Trading Price Based on VAR and ARIMA-LSTM Model
WEI Jing
LIU Furan
YUAN Ting
WANG Ting
Abstract：To analyze and predict the influencing factors of carbon emission trading prices,the Hubei carbon emission trading market was taken as the research object,and 11 influencing factors from five aspects(macroeconomics,energy prices,exchange rates,international carbon markets,and climate environment)were used for sorting and correlation analysis.At the same time,a vector autoregression(VAR)model was established for empirical analysis,and further,autoregressive integrated moving average(ARIMA)model,long short term memory(LSTM),and weighted combination models were further used for comparative prediction.The research found that:Carbon trading price fluctuations were mainly driven by their own historical prices,with a contribution rate of up to 98.07%;The prediction accuracy of the ARIMA-LSTM weighted combination model was significantly improved compared with the single model,with mean absolute percentage error(MAPE)reduced to 1.943 and root mean square error(RMSE)reduced to 0.915;The current carbon market in Hubei had problems such as an incomplete trading mechanism and a single price formation mechanism,which posed potential systemic risks.The research confirmed that it was urgent to improve the operational efficiency of the carbon market through measures such as improving market infrastructure construction and optimizing price formation mechanisms.The research results could provide methodological support for regulatory authorities to build a risk warning system and providing quantitative decision-making basis for climate policy formulation.
Keywords：carbon emission trading priceinfluencing factorsVAR modelARIMA modelLSTM model
Publication Date：2025-06-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：10( 280-289 )