论文部分内容阅读
组合预测理论及建模技术对于信息不完备的复杂经济系统具有一定的实用性。鉴于能源消费系统的复杂性及非线性的特征,文章首先利用陕西能源消费量的历史数据,分别采用指数回归模型、能源需求弹性回归模型及灰色模型建立了陕西省能源消费系统的单项预测模型。其次,采用标准差法进行非负权重分配,建立了陕西省能源消费量的组合预测模型。结果表明,加入时间虚拟变量和分段建模的预测精度明显提高,且组合预测模型的精度高于单项预测模型。最后,应用该模型对陕西未来10年的能源消费量进行了预测。
Combination forecasting theory and modeling technology have some practicality for complex economic system with incomplete information. In view of the complexity and non-linear characteristics of energy consumption system, the article first uses the historical data of energy consumption in Shaanxi Province to establish a single-item prediction model of energy consumption system in Shaanxi Province using exponential regression model, energy demand elasticity regression model and gray model respectively. Secondly, using the standard deviation method for non-negative weight distribution, a combined forecasting model of energy consumption in Shaanxi Province was established. The results show that the prediction accuracy of time-varying dummy variables and piecewise modeling are obviously improved, and the accuracy of combined forecasting model is higher than that of single forecasting model. Finally, the model is used to predict the energy consumption of Shaanxi in the next 10 years.