基于变异系数的DEA交叉效率风险型决策方法  

Risk Decision Method for DEA Cross-efficiency Based on Variable Coefficient

作　　者：王慧颖 程幼明 Wang Huiying;Cheng Youming(College of Management Engineering,Anhui Polytechnic University,Wuhu Anhui 241000,China)

机构地区：[1]安徽工程大学管理工程学院,安徽芜湖241000

出　　处：《统计与决策》2020年第22期160-164,共5页Statistics & Decision

基　　金：国家自然科学基金资助项目(71671001,71801003);安徽工程大学研究生实践与创新项目。

摘　　要：在现行DEA交叉效率评价过程中,效益集结时通常对行向量信息采用均值集结显然是一种典型的不确定型等概率原则,导致交叉矩阵中决策信息大量遗失且结果缺乏区分度,从而使得此评价缺乏客观性和全面性。文章提出不仅考虑交叉矩阵中行向量信息,更要考虑列向量决策信息,通过定义行向量信息为决策方案的状态,定义列向量信息为状态出现的概率,从而充分利用交叉矩阵中的决策信息。设计出基于变异系数的DEA交叉效率评价模型,提出DEA交叉效率风险型决策方法,相对于经典DEA交叉效率评价法更加简洁直观,同时具有理论依据,最后通过算例验证了算法的合理性和有效性。In the current DEA(data envelopment analysis)crossover efficiency evaluation process,it is obviously a typical probability principle of uncertain type to use mean aggregation for row vector information in benefit aggregation,which leads to massive loss of decision information in the crossover matrix and the result lacking in differentiation degree,thus causing the evaluation to lack objectivity and comprehensiveness.Addressing this problem,this paper proposes the solution of considering not only the row vector information in the cross-matrix,but also the column vector decision information.By defining the row vector information as the state of the decision scheme and the column vector information as the probability of state occurrence,the paper makes full use of the decision information in the cross matrix.The paper also designs a DEA cross-efficiency evaluation model based on variable coefficient,and presents the DEA cross-efficiency risk decision method,which,compared with the classical DEA cross-efficiency evaluation method,is more concise and intuitive with theoretical basis.Finally,the paper gives an example to verify the rationality and effectiveness of the algorithm.

关 键 词：DEA 交叉效率 变异系数 风险型决策 

分 类 号：C934[经济管理—管理学]