Greenhouse gas emissions and road infrastructure in Europe: A machine learning analysis
本文结合计量模型与机器学习，分析2013-2021年欧洲交通变量对温室气体排放的影响，发现汽车密度增加排放，而替代燃料车辆和公共交通（尤其有轨电车）能显著减排，为政策制定者提供推广电动车和改善基础设施的启示。
• GHG emissions in Europe dropped significantly from 2013 to 2021, with regional disparities. • Higher car density (PC1000) correlates with increased GHG emissions due to more fuel use. • Alternative fuel vehicles, like mopeds and motorcycles, significantly reduce emissions. • Public transportation systems, especially trams, are crucial in lowering urban GHG emissions. • Policy implications suggest promoting electric vehicles and improving public transport infrastructure. This paper explores the determinants of greenhouse gas (GHG) emissions in Europe, focusing on transportation-related variables. By combining classical econometric models with Machine Learning (ML) techniques, we analyze data spanning from 2013 to 2021. The empirical findings highlight the complex relationship between newer passenger cars and GHG emissions, noting the significant impact of their production and increased usage. Conversely, the adoption of alternative fuel vehicles is found to significantly reduce emissions. This is further supported by ML models, which emphasize the critical role of car density and alternative fuel vehicles in determining emissions. Policy implications suggest the need for targeted interventions, including the promotion of electric and hybrid vehicles, enhancements in transportation infrastructure, and the implementation of economic incentives for clean technologies.