Forecasting China's per capita energy consumption using dynamic grey model based on ARIMA model

Energy is the cornerstone of social progress, and energy consumption is an important reference data when the government makes energy policies. Therefore, prediction of energy consumption can provide a basis for the formulation of policies. In the traditional grey system theory, the grey input is always regarded as an invariant after determined by the least square method. However, the grey input would be changed along with the change of the internal information in the grey system. As a result, the performance and accuracy of the grey model would be affected by it. Therefore, this paper proposes the ARIMA-b-GM (1,1) model, which uses the ARIMA prediction model to fit and predict the grey input and realize the dynamic change of the grey input. Then the time response sequence of the ARIMA-b-GM (1,1) model is deduced by changing the form of grey differential equation. Finally, the ARIMA-b-GM (1,1) model is applied to predict the per capita energy consumption in China. The results exhibit that ARIMA-b-GM (1,1) model has higher prediction accuracy than GM (1,1) model, NGM (1,1) model and SAIGM (1,1) model.