Exploring carbon dioxide emissions forecasting in China: A policy-oriented perspective using projection pursuit regression and machine learning models

• Published in: Technological forecasting & social change Vol. 197; p. 122872
• Main Authors: Chang, Lei, Mohsin, Muhammad, Hasnaoui, Amir, Taghizadeh-Hesary, Farhad
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01-12-2023
• Subjects: Emission intensity; Forecasting carbon emission; Machine learning model; Machine learning model; Emission intensity; Forecasting carbon emission
• ISSN: 0040-1625; 1873-5509
• DOI: 10.1016/j.techfore.2023.122872

Achieving a balance between future greenhouse gas reduction and sustained economic growth is of utmost importance. This study leverages machine learning (ML), specifically projection pursuit regression (PPR), to evaluate the key factors that influence CO2 emission predictions in China. The analysis notably identifies the escalating electricity consumption as a primary influencing factor. Based on empirical findings, it is evident that building electricity consumption will continue to rise steadily until 2050 unless new restrictions or technological advancements are implemented. Relying solely on the reduced carbon intensity of electricity will not enable China to achieve carbon neutrality. Therefore, there is a pressing need for more energy-efficient building retrofits and technologies to reduce power consumption in both residential and commercial properties. This policy-oriented study underscores its practical implications, offering valuable insights to policymakers for developing targeted CO2 reduction strategies that align with sustainable development and climate goals.
•Machine learning assesses CO2 factors in China.•Rising power usage: key influence•Building retrofits for carbon neutrality•Policy-driven CO2 reduction with machine-learning