Exploring the driving forces of distributed energy resources in China: Using a semiparametric regression model

Abstract Developing distributed energy resources can help reduce carbon dioxide emissions and control environmental pollution. Investigating the main driving factors of distributed energy resources can provide empirical support for government departments to formulate relevant energy policies. Different from traditional linear models, the semiparametric regression model has data-driven characteristics and can reveal possible nonlinear relationships between economic variables. Based on 2005–2017 panel data, this article uses the semiparametric regression model to investigate distributed energy resources in China. Estimated results show that technological progress has the largest impact on the distributed energy resources in the western region, due to the difference in R&D expenditures and patented technology. Foreign oil dependence produces a greater effect on the distributed energy resources in the eastern region, because it imports the most oil. The impact of energy subsidies in the central and western regions is greater, because their financial subsidies for renewable energy and natural gas consumption have grown faster. Industrial structure has an inverted N-shaped nonlinear impact in the eastern region, but exerts a positive U-shaped impact in the central and western regions. In addition, distributed energy resources have significant geographic differences. In the future, we should take the spatial element into consideration when exploring distributed energy resources.