基于VAR和GM组合模型的电力消费预测

Electricity Consumption Forecast: Based on VAR and GM Combination Model

摘要: 组合预测模型较单一模型具有更准确的预测精度,利用中国电力消费1978年～2007年的年度数据,将VAR模型与灰色预测的GM(1,1)、无偏GM(1,1)两种模型有机组合,建立灰色VAR组合预测模型,对2008年～2015年的电力消费量进行预测。结果表明,该模型平均相对误差为4.88％,拟合精度较高,具有较好的应用价值。

Abstract: A combined forecast model is more accurate than a single forecasting model. Based on the electricity consumption data from 1978 to 2007, by combing the VAR model, the GM(1,1) model and the WGM(1,1) model, we set up the grey VAR combined forecast model and forecast the electricity consumption from 2008 to 2015. The research results show that the forecast values are accurate: the average relative error is only 4.88%, so it can be used in the application as one of the efficient methods.