Prioritizing driving factors of household carbon emissions: An application of the LASSO model with survey data

Abstract

Abstract The identification of factors that influence household carbon emissions (HCEs)—a key driver of the national emissions, is an important step in achieving more accurate predictions, as well as better interpretation and more effective policy intervention. In this paper, based on survey data, we first calculated the direct, indirect, and total HCEs per capita for 37,620 households in China in the year of 2012, 2014 and 2016. Then we introduced a LASSO regression model to determine the main driving factors of HCEs and ranked the factors according to their importance. The use of the LASSO regression model addresses the issues of multicollinearity and over-fitting. It also provides two practical benefits: minimizing the number of influencing factors for forecasting and giving more flexibility in policy design. The results showed that fuel type and dwelling type can explain more than 70% of the direct HCEs, while income, urban or rural residency, and fuel type are the three most important influencing factors of the indirect HCEs. To mitigate HCEs while China will continue its rapid urbanization and fast consumption growth, the government needs to provide affordable clean energy, improve the efficiency of household energy consumption, promote green and low-carbon economic recovery, and guide low-carbon lifestyles.