趋势外推与ARMA组合的能源需求预测模型

能源需求预测是制定能源规划的重要组成部分.鉴于能源需求系统的复杂性和非线性特征,文章结合我国能源需求的历史数据,首先用趋势外推预测建立了单项预测模型,并对趋势外推预测的优缺点进行了分析;然后运用趋势外推与ARMA组合模型对趋势外推模型中的非趋势分量即残差序列进行分析,得到趋势外推与ARMA组合模型的预测结果;最后通过比较得出趋势外推与ARMA组合模型的预测结果不仅准确性高、而且稳定性好这一结论.这对于我国能源需求预测有着重要的理论意义和现实意义.

An Energy Demand Prediction Model Based on Trend Extrapolation and ARMA

Prediction of energy demand is a significant component in making energy plan.In view of the complexity and non-linear characteristics of energy demand system,this paper,by making use of the historical data about energy demand,first establishes a prediction model based on trend extrapolation alone and analyses its merits and demerits.Then,an analysis is made of the non-trend component or deviational variance of the trend extrapolation model by using the combination of trend extrapolation and ARMA,arriving at the result of the combined model.Finally,a comparison of the results is made,which shows that the prediction result of the combined model based on trend extrapolation and ARMA is not only more accurate,but also more stable.Thus the combined model is of great theoretical and practical significance to China's energy demand prediction.