Economic growth, energy consumption, and urbanization are mutually influenced, therefore, forecasting economic growth, energy consumption and urbanization in China has always been important as it could help government to improve energy policies and plans. To this end, two novel grey multivariable models, SRMGM(1, M) and BRMGM(1, m), are designed in this paper. Compared with the MGM(1, M) model, the SRMGM(1, M) model can optimize the background of the traditional model according to the Simpson formula, and the BRMGM(1, m) model optimizes the background value according to a Boolean formula. The reconstruction of background value can reduce the prediction error of MGM(1, M) model. Furthermore, rolling prediction is added to these two new models based on the priority principle of new information. In order to validate their efficacy and accuracy, the two proposed models are employed to reproduce and predict economic growth, energy consumption, and urbanization compared with MGM(1, m) model and three linear regression models. FDs of our models, in both fitted and predicted stages are all over 97%, and the ADFs are all below 5%. The results show that our new models achieves higher fitted and predicted accuracy than MGM(1,m) model and the non-grey models. Eventually, the new models will be used to forecast economic growth, energy consumption, and urbanization from 2020 to 2025, and the forecast results provide a solid basis for economic growth policies and energy consumption plans.
Read full abstract