摘要
主要指出文献[1],[2]中所用的"新方法"不能完全区分决策单元的DEA有效性和弱DEA有效性.同时,"新方法"中所使用的DEA模型(即文献[3]中超效率DEA模型)的最优解不一定存在,这也是"新方法"使用中的一大缺陷.本文同时指出"新方法"虽然是可以扩充的,但扩充后,某些"新模型"仍然会出现上述问题.如果单纯的去评价决策单元的DEA有效性、弱DEA有效性和非弱DEA有效性时,建议还是使用传统的经典模型为好;如果需要进一步对DEA有效性再进行分析,是可以象最早提出超效率DEA模型的文献[3]中那样去应用超效率DEA模型。
This article indicates that the new method in references [1] and [2] can not completely discriminate the DEA efficiency and weak DEA efficiency of DMUs, furthermore, the optimal solution of DEA models used in the new method might not exist, which demonstrates the limitation of the application of the new method. The article also points out that although the new method could be extended,the abovementioned problems might still exist after the extension. It is proposed that the traditional models are the best when the DEA efficiency, weak DEA efficiency or not weak DEA efficiency of DMUs are simply evaluated, and if further analysis on DEA efficiency is required, the super efficiency DEA models which are established in reference [3] can be applied.
出处
《数学的实践与认识》
CSCD
北大核心
2007年第22期93-97,共5页
Mathematics in Practice and Theory
基金
国家自然科学基金(NNSF70531040NNSF70371058)
中国人民大学985工程"中国经济研究哲学社会科学创新基地"
