Influencing Factors And Prediction Analysis Of Carbon Emission Under The Background Of Digital Economy

Nowadays,our country is entering the new era of digital economy,and low carbon,green,digital is major trend of social progress in the future.As a new form of economic and social development,the digital economy has deeply influenced people’s production and life,changed the employment structure,and spawned a huge adjustment of our industrial structure.The application of new technologies and new products derived from the digital economy in the energy system can not only save energy,reduce emissions and improve energy efficiency,but also stimulate carbon reduction and even negative carbon technological innovation.Digital economy has become an important starting point of carbon peaking and carbon neutrality goals,so exploring the impact of digital economy on carbon emission has great influence on the development of green economy.Starting from the concept of digital economy and on the basis of the existing achievements of many scholars,this article uses the sample data from nine provinces and regions of the Yellow River Economic Belt,covering the period from 2013 to2019,the index system for evaluating the development level of digital economy in the Yellow River Economic Belt has been set up,and the combined method of global principal component analysis and global entropy is adopted.The indicators of the development level of digital economy were weighted and calculated,and then a fixed effect model based on panel data was constructed to explore the relationship between digital economy and carbon emissions,and the robustness of the empirical results was tested.In addition,this paper also uses machine learning algorithms,the random forest model,standard BP neural network model and PSO-BP neural network model were constructed to predict and analyze the carbon emissions in the Yellow River Economic Belt,and the models were compared and evaluated based on MAE,MSE,RMSE and R~2.The empirical results of this paper illustrate that:(1)The progress of digital economy in the Yellow River Economic Belt is significantly different and uneven among regions;(2)Advances in the digital economy in these regions has an inhibitory effect on carbon emissions,and the results is still useful after the robustness test;(3)The performance of BP neural network model optimized by PSO algorithm is superior to tranditional network and random forest algorithm in each evaluation index and precision of prediction,and MAE,MSE,RMSE and R~2are 17.0700,517.4782,22.7481 and 0.99364,respectively.Compared with standard BP neural network,MAE,MSE and RMSE predicted by PSO-BP neural network decreased by 71.57%,91.54%and 70.91%,respectively.Based on the research conclusions of this paper,the following suggestions are put forward:(1)Make in-depth research to understand the reasons for the relatively lagging and uneven development of the digital economy,combine the development status of different regions,and improve the digital economy in Yellow River Economic Belt according to local conditions;(2)Seize the opportunity of digital development,increase investment in digital infrastructure,and improve regional infrastructure construction;In order to promote the long-term and sound development of the digital industry,it is necessary to transform traditional industries into green and low-carbon industries,and achieve mutual benefit and win-win results of economic and environmental benefits;(3)Increase the investment in education in the Yellow River Economic Belt,improve the overall quality of residents,and narrow the human capital gap between regions.