基于信息含量最大的绿色产业评价指标筛选模型及应用

Green industry evaluation indicators screening model based on the maximum information content and its application

绿色产业评价是发展低碳经济,实现经济、社会与生态环境可持续发展的必要环节,建立合理的指标体系是绿色产业评价的关键.从反映绿色产业内涵的角度出发,从绿色生产、绿色消费、绿色环境三个方面海选指标,依据信息含量最大和冗余信息剔除原则定量筛选指标,构建了绿色产业评价指标体系.文章的创新与特色一是通过变异系数求解指标信息含量,遴选信息含量最大的指标.构建了由绿色生产、绿色消费、绿色环境三个准则层26个指标组成的绿色产业评价指标体系.二是通过变异系数筛选出各个类中变异系数(信息含量)最大的指标,保证了筛选出的指标对绿色产业评价有显著影响.三是通过R聚类将同一准则层内的指标分类,使不同类别的指标反映不同的数据特征.避免了绿色产业评价中不同指标反映的信息重复.四是研究表明:最终建立的指标体系用27.7%的指标反映了98.44%的原始信息.

Green industry evaluation is a necessary part to develop low-carbon economy and achievingsustainable development in the economic, social and ecological environment. Establishing a reasonableindicators system is the key to green industry evaluation. From the connotation of green industry, this paperchooses green industry evaluation indicators extensively which include green production, green consumptionand green environment. According to the principle of the maximum information content and the principleof eliminating redundant information, it establishes the evaluation indicators system of green industry.The special and contributions of this paper lie on four aspects. Firstly, it uses coefficient of variation tocalculate the information content of each indicator, then the indicator with the highest information contentis screened. At last, the evaluation indicators system of green industry composed of three criterion layersgreen production, green consumption and green environment, which includes 26 indicators, is established.Secondly, it uses coefficient of variation to screen the indicator with the highest information content loadingin each cluster, which ensures that the selected indicator has the greatest influence on the green industryevaluation. Thirdly, it uses R cluster analysis to classify indicators in the same criterion layer, whichensures that different indicators represent the different characteristics of the data and avoids the differentindicators reflect repeated information. Fourthly, the study result reveals that the final indicators systemreflects 98.44% of original information by 27.7% of initial indicators.