Scholarlink

<p>Enhancing environmental sustainability is a major objective, and various econometric approaches have been used to understand it. In this regard, this paper aims to build upon the current literature in environmental economics by examining the relationships among energy consumption, economic growth, renewable energy, and carbon dioxide (CO<sub>2</sub>) emissions in Pakistan during the period 1965–2022 using machine learning (ML) based methodologies and the traditional autoregressive distributed lag (ARDL) model. Based on an analysis of data using novel machine learning methods, it has been determined that reducing CO<sub>2</sub> requires a comprehensive shift from fossil-fuel-based energy sources to renewable alternatives, alongside adopting a more sustainable trajectory. The application of ML-based regression and prediction techniques indicates a projected decrease in CO<sub>2</sub> for the year 2023. However, from 2024 onward, findings show a rise in CO<sub>2</sub> emissions driven by increased use of non-renewable energy sources. On the other hand, the ARDL results indicate that energy use has a positive impact on CO<sub>2</sub> emissions. In contrast, economic growth has a significant negative impact on CO<sub>2</sub> emissions in both the short- and long-run. In addition, the impact of renewable energy is only short-term and significant in reducing CO<sub>2</sub> emissions. Thus, providing Pakistan with a consistent, sustainable economic development trajectory through renewable energy is crucial.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Harnessing machine learning to decode the energy–economy–environment trilemma: insights from renewable and non-renewable energy use in Pakistan

  • 期刊论文

  • 发表日期：

    2026年7月21日

查看全文
  • Muhammad Qayyum,
  • Minhaj Ali,
  • Muhammad Waleed Hassan,
  • Usman Mehmood

摘要

Enhancing environmental sustainability is a major objective, and various econometric approaches have been used to understand it. In this regard, this paper aims to build upon the current literature in environmental economics by examining the relationships among energy consumption, economic growth, renewable energy, and carbon dioxide (CO2) emissions in Pakistan during the period 1965–2022 using machine learning (ML) based methodologies and the traditional autoregressive distributed lag (ARDL) model. Based on an analysis of data using novel machine learning methods, it has been determined that reducing CO2 requires a comprehensive shift from fossil-fuel-based energy sources to renewable alternatives, alongside adopting a more sustainable trajectory. The application of ML-based regression and prediction techniques indicates a projected decrease in CO2 for the year 2023. However, from 2024 onward, findings show a rise in CO2 emissions driven by increased use of non-renewable energy sources. On the other hand, the ARDL results indicate that energy use has a positive impact on CO2 emissions. In contrast, economic growth has a significant negative impact on CO2 emissions in both the short- and long-run. In addition, the impact of renewable energy is only short-term and significant in reducing CO2 emissions. Thus, providing Pakistan with a consistent, sustainable economic development trajectory through renewable energy is crucial.