中国能源消费的区域特征与相关性分析

Regional characteristics of Chinese energy consumption and correlation analysis

为研究能源消费的区域特征和相关性,根据1991—2000年、2001—2010年中国30个地区的能源消费和人均GDP面板数据,考虑不同省份产业结构的差异性,对能源消费评价指标进行了修正并分解为强度效应、结构效应和收入效应。运用KMeans面板聚类方法将30个地区划分为4种不同的能源消费区域:后工业化区、高度工业化区、资源能源主导区和起步发展区;后工业化区结构效应和收入效应最大,能源利用效率较高;高度工业化区为东部沿海区域,结构效应和收入效应较大,能源消费强度效应最小,能源利用效率最高;资源能源主导区和起步发展区为中西部地区,结构效应和收入效应均较小,能源消费强度效应较大,能源利用效率较低。运用贝叶斯面板数据推断方法分析区域能源消费相关性的动态变化。研究结果表明,后工业化区与高度工业化区、高度工业化区与起步发展区的能源消费相关性均呈明显增强趋势。

In order to research the regional characteristics and correlation of energy consumption, according to energy consumption and panel data of actual GDP per capita from China＇s 30 regions in 1991-2000 and 2001-2010, and taking into account differences of industrial structure in various provinces,the author corrected energy consumption evaluation indicator and broken it down into energy consumption intensity effect, structure effect and income effect. Using K-Means plane clustering method,China is divided into four energy consumption regions： post-industrial region, high industrialized region, resources and energy economy oriented region and primary development region. Of them, the post-industrial region has the maximum structural effect and income effect, with higher energy efficiency; highly industrialized region is concentrated in the eastern coastal region, with larger structure and income effects and the minimum energy consumption intensity effect and the highest energy efficiency; while for the latter two categories, they are in central and western regions, with smaller structure and income effects and lower energy efficiency. Additionally,Bayesian panel data inference method is also adopted to analyze dynamic changes to regional energy consumption correlation. The results reveal that： energy consumption correlations between post-industrial area and high industrialized coastal area, and between high industrialized coastal area and primary development area both show a significantly upward trend.