{"title":"Factors affecting per capita ecological footprint in OECD countries: Evidence from machine learning techniques","authors":"Muhammed Sehid Gorus, E. Karagol","doi":"10.1177/0958305X221112913","DOIUrl":null,"url":null,"abstract":"For a few decades, factors affecting environmental deterioration have been at the center of much interest This paper examines the impact of income level, disaggregated energy consumption, types of globalization level, and urbanization on per capita ecological footprint by utilizing novel machine learning techniques (tree regression, boosting, bagging, and random forest) for 27 OECD countries during 1971–2016. It is found that the random forest algorithms best fit the dataset. The empirical results exhibit that oil product consumption, electricity consumption, and gross domestic product are the most significant variables for our model. Besides, the partial dependence plots results show that economic growth and especially fossil fuel energy consumption damage the environment. These findings have important implications for both developed and developing countries for designing proper energy and environmental policies. Especially, policymakers should focus on sustainable development instead of plain economic growth.","PeriodicalId":11652,"journal":{"name":"Energy & Environment","volume":"21 1","pages":"2601 - 2618"},"PeriodicalIF":4.0000,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Energy & Environment","FirstCategoryId":"93","ListUrlMain":"https://doi.org/10.1177/0958305X221112913","RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENVIRONMENTAL STUDIES","Score":null,"Total":0}
引用次数: 0
Abstract
For a few decades, factors affecting environmental deterioration have been at the center of much interest This paper examines the impact of income level, disaggregated energy consumption, types of globalization level, and urbanization on per capita ecological footprint by utilizing novel machine learning techniques (tree regression, boosting, bagging, and random forest) for 27 OECD countries during 1971–2016. It is found that the random forest algorithms best fit the dataset. The empirical results exhibit that oil product consumption, electricity consumption, and gross domestic product are the most significant variables for our model. Besides, the partial dependence plots results show that economic growth and especially fossil fuel energy consumption damage the environment. These findings have important implications for both developed and developing countries for designing proper energy and environmental policies. Especially, policymakers should focus on sustainable development instead of plain economic growth.
期刊介绍:
Energy & Environment is an interdisciplinary journal inviting energy policy analysts, natural scientists and engineers, as well as lawyers and economists to contribute to mutual understanding and learning, believing that better communication between experts will enhance the quality of policy, advance social well-being and help to reduce conflict. The journal encourages dialogue between the social sciences as energy demand and supply are observed and analysed with reference to politics of policy-making and implementation. The rapidly evolving social and environmental impacts of energy supply, transport, production and use at all levels require contribution from many disciplines if policy is to be effective. In particular E & E invite contributions from the study of policy delivery, ultimately more important than policy formation. The geopolitics of energy are also important, as are the impacts of environmental regulations and advancing technologies on national and local politics, and even global energy politics. Energy & Environment is a forum for constructive, professional information sharing, as well as debate across disciplines and professions, including the financial sector. Mathematical articles are outside the scope of Energy & Environment. The broader policy implications of submitted research should be addressed and environmental implications, not just emission quantities, be discussed with reference to scientific assumptions. This applies especially to technical papers based on arguments suggested by other disciplines, funding bodies or directly by policy-makers.