基于前景理论的交叉效率集结方法

陈磊,王应明

系统科学与数学 ›› 2018, Vol. 38 ›› Issue (11) : 1307-1316.

PDF(467 KB)
PDF(467 KB)
系统科学与数学 ›› 2018, Vol. 38 ›› Issue (11) : 1307-1316. DOI: 10.12341/jssms13490
论文

基于前景理论的交叉效率集结方法

    陈磊,王应明
作者信息 +

Cross-Efficiency Aggregation Method Based on Prospect Theory

    CHEN Lei ,WANG Yingming
Author information +
文章历史 +

摘要

传统交叉效率方法往往采用相加平均的方式来集结效率, 这不仅缺乏足够的理 论依据, 而且导致大量决策信息的遗失. 针对这个问题, 文章引入前景理论来研究决策者 面临收益和损失时的主观价值感受, 并分别在乐观型、中立型和悲观型决策偏好的引导下构建相应的效率集结方法. 随后, 引入距离熵的概念来衡量不同决策单元视角下评价结果的可靠性, 以此修正交叉效率集结结果. 该集结方法充分考虑了决策者的主观偏好, 并在其引导下最大程度地保留了决策信息, 从而获得最符合现实决策需求的效率评价结果. 最后, 通过案例分析来验证该方法的有效性.

Abstract

The additive mean method is always applied to aggregate efficiency in the traditional cross efficiency method. However, this method not only lacks sufficient theoretical basis, but also leads to the loss of a lot of decision information. Therefore, in this paper, the prospect theory is introduced to study the subjective value perception of the decision-makers when they face the benefits and losses; and then the efficiency aggregation methods are constructed based on the guidance of optimistic, neutral and pessimistic decision preference, respectively. Furthermore, the concept of distance entropy is introduced to measure the reliability of the evaluation results from the perspective of different decision-making units, so as to modify the results of the cross efficiency aggregation. The aggregation methods proposed in this paper take the subjective preferences of decision-makers into full consideration, and preserve decision information to the largest extent, so as to get the most effective evaluation results that meet the needs of realistic decision-making. Finally, the effectiveness of these methods is verified by case analysis.

关键词

数据包络分析 / 决策偏好 / 前景价值 / 距离熵 / 交叉效率集结.

引用本文

导出引用
陈磊 , 王应明. 基于前景理论的交叉效率集结方法. 系统科学与数学, 2018, 38(11): 1307-1316. https://doi.org/10.12341/jssms13490
CHEN Lei , WANG Yingming. Cross-Efficiency Aggregation Method Based on Prospect Theory. Journal of Systems Science and Mathematical Sciences, 2018, 38(11): 1307-1316 https://doi.org/10.12341/jssms13490
PDF(467 KB)

430

Accesses

0

Citation

Detail

段落导航
相关文章

/