可处理负数据的超效率数据包络分析模型

An Analytical Model of Super-efficiency Data Envelopment in the Presence of Negative Data

基于方向距离函数,通过选择恰当的方向向量,构造了全新的可处理负数据的超效率模型.与已有相关模型相比,该模型不但能够对所有DMU进行完全排序,而且无论投入产出数据是正是负,它的解都是可行的.理论分析和实证结果证明,新模型成功地解决了传统超效率模型的不可行问题,并且在处理负数据时总是可行的.

A new super-efficiency model based on the directional distance function to handle the negative data by means of the selection of a proper direction vector is constructed in this paper. Compared with existing related models, the proposed model can rank the efficient DMUs completely and no matter whether the input-output data are positive or negative, its solution is always feasible. Theoretical analysis and empirical results demonstrate that the new model solves the infeasibility issue of the conventional super-efficiency model successfully and is always feasible for dealing with negative data.