改进的公共权重的区间DEA效率评价  

Improved Public Weights for Interval DEA Efficiency Evaluation

作　　者：孙雪莲 郭良栋[1] SUN Xue-lian;GUO Liang-dong(School of Science,University of Science and Technology Liaoning,Anshan 114000,China)

机构地区：[1]辽宁科技大学理学院,辽宁鞍山114000

出　　处：《数学的实践与认识》2023年第10期106-115,共10页Mathematics in Practice and Theory

摘　　要：提出了一种改进的公共权重的考虑非期望产出的区间数据包络模型(DEA)评价模型.通过Pearson相关性检验,VIF检验法和容忍度检验法对投入,期望产出和非期望产出指标进行约简,并计算出指标的客观权重,引入调和参数使客观权重满足DEA模型限制条件且存在最优解.该模型有效解决了传统DEA模型中可能发生的公共权重分配为零的情况发生,同时考虑了指标负面评价的情况,增加评价结果的区分度.对中国北方十个省份的矿业环境治理效率进行了实证分析,构建了矿山生态环境治理效率评价体系,与相关文献进行比较,表明该模型的有效性且可以有效降低计算复杂度,提高计算效率.This paper presents an improved evaluation model of interval Data Enveloping model(DEA)which takes into account undesired outputs with common weights.Pearson correlation test,VIF test and tolerance test were used to reduce the input,expected output and non-expected output indicators,and the objective weights of the indicators were calculated.Harmonic parameters were introduced to make the objective weights meet the constraints of DEA model and there are optimal solutions.This model effectively solves the possible situation that the common weight distribution is zero in the traditional DEA model,and at the same time considers the negative evaluation of indicators to increase the differentiation of evaluation results.Based on the empirical analysis of mining environmental governance eficiency in ten provinces in northern China,an evaluation system of mine ecological environmental governance efficiency was established.The comparison with relevant literature shows that the model is effective and can effectively reduce the computational complexity and improve the computational eficiency.

关 键 词：公共权重区间DEA 权重约束 效率评价 矿业环境治理 

分 类 号：F223[经济管理—国民经济]