中国能源需求预测函数：主成份辅助的协整分析

Predictive Function of Energy Demand in China: Co-integration Analysis Assisted by Principal Components Analysis

本文运用主成份分析法辅助协整和误差校正模型建立中国能源需求的预测模型,并利用数据进行实证检验,结果表明中国能源价格已经能在一定程度上反映能源供求状况,而且利用产业结构调整降低能耗的方法是可行的。在长期中,人口结构,产业结构,燃料价格、消费水平、矿物燃料、润滑油及有关原料的进出口水平影响着中国的能源需求;在短期中,经济产出和产业结构灵敏度较高,显著影响能源需求。该模型预测效果良好,可以为能源需求的预测与管理提供参考。

This paper uses co-integration analysis and Error-Correction Model（ECM） assisted by principal components analysis to construct predictive function of energy demand in China, and test the function with data, results show that： energy price can reflect the situation of supply and demand of energy, and the method that reduce energy consume by adjust the industrial structure is available. In long term, population structure, industrial structure, consume level, fuel＇s import and export influent energy demand in China; in short term, the sensitivity of GDP and industrial structure is higher. Predictive effect of the function is so excellent that we can infer it when plan energy strategy.