Dynamic Decomposition Analysis and Forecasting of Energy Consumption in Shanxi Province Based on VAR and GM (1, 1) Models

Abstract

Environmental issues caused by energy consumption have attracted increasing attention recently. Shanxi Province, a typical energy-dominated region in China, has long-term dependency on coal industry generating extensive economic growth, which is detrimental to green development. Distinguished from previous studies ignoring driving factors of energy consumption, this paper establishes a vector autoregression (VAR) model to dynamically identify the drivers of energy consumption based on STIRPAT model in Shanxi Province from 1990 to 2015. It can be obtained from the impulse response analysis that a positive shock in population, GDP, and urbanization level, respectively, positively affect energy consumption, and a positive change in technology negatively affects energy consumption in the long run. The variance decomposition results indicate that fluctuation in energy consumption explained by the innovation of the urbanization level accounts for 23.18%, which plays a prevailing role in increasing energy consumption. Meanwhile, the forecasting results of GM (1,1) model manifest that energy consumption in Shanxi Province generally has an increasing trend from 2016 to 2025. Consequently, Shanxi can achieve green development through optimizing energy structure, promoting the transformation of resource-based cities, and promoting low-carbon technological innovation. This paper can be available for other resource-based regions analogous to Shanxi.