scholarly journals Grey Relational Analysis Method for Group Decision Making in Credit Risk Analysis

Author(s):  
Wenshuai Wu
Filomat ◽  
2016 ◽  
Vol 30 (15) ◽  
pp. 4135-4150 ◽  
Author(s):  
Wenshuai Wu ◽  
Gang Kou ◽  
Yi Peng

Credit risk analysis is a core research issue in the field of financial risk management. This paper first investigates the analytic hierarchy process (AHP) as a method of measuring index weights for group decision-making (GDM). AHP for group decision-making (AHP-GDM) is then researched and applied, taking into full account the cognitive levels of different experts. Second, the concept of grey relational degree is introduced into the ideal solution of the technique for order of preference by similarity to ideal solution (TOPSIS). This concept fully considers the relative closeness of grey relational degree between alternatives and the ?ideal? solution in order to strengthen their relationship. The AHP-GDM method overcomes the problem of subjectivity in measuring index weights, and the revised TOPSIS (R-TOPSIS) method heightens the effectiveness of assessment results. An illustrative case using data from Chinese listed commercial banks shows that the R-TOPSIS method is more effective than both TOPSIS and grey relational analysis (GRA) in credit risk evaluation. The two improved multi-criteria decision making (MCDM) methods are also applied to empirical research regarding the credit risk analysis of Chinese urban commercial banks. The results indicate the validity and effectiveness of both methods.


2021 ◽  
Author(s):  
Decai Sun ◽  
Dang Luo

Abstract For the uncertainty and complexity ingroup decision making and the differences of decision makers’ reliabilities, a group decision making method based on grey relational analysis and evidence theory is proposed. Combining grey relational analysis with evidence theory, a novel decision-making method extracting the degree of ignorance for individual decision makers’ information and constructing the Mass function is presented based on the comprehensive grey relational analysis (CGRA) method. We should also address how AI systems make their black box decisions, which calls for research into Explainable AI (XAI) by pursuing reverse engineering and self-explainability in AI. Considering the differences of decision makers’ reliabilities, the Mass function is modified by the evidence weight, and the group decision information is fused by the Dempster’s combination rule. On this basis, the Mass function is further transformed into the probability by the Pignistic probability transformation, which issued for ranking analysis of group decision making. Finally, the proposed method is applied to the green supplier selection, and the comparative analysis is further performed to verify the rationality and effectiveness of the proposed method.


2021 ◽  
Vol 41 (2) ◽  
pp. 3783-3795
Author(s):  
Shanshan Zhang ◽  
Hui Gao ◽  
Guiwu Wei ◽  
Xudong Chen

The Multi-attribute group decision making (MAGDM) problem is an interesting everyday problem full of complexity and ambiguity. As an extended form of fuzzy sets, intuitionistic fuzzy sets (IFSs) can provide decision-makers (DMs) with a wider range of preferences for MAGDM. The grey relational analysis (GRA) is an effective method for dealing with MAGDM problems. However, in view of the incomplete and asymmetric information and the influence of DMs’ psychological factors on the decision result, we develop a new model that GRA method based on cumulative prospect theory (CPT) under the intuitionistic fuzzy environment. Moreover, the weight of attribute is calculated by entropy weight, so as to distinguish the importance level of attributes, which greatly improves the credibility of the selected scheme. simultaneously, the proposed method is used to the selection of optimal green suppliers for testifying the availability of this new model and the final comparison between this new method and the existing methods further verify the reliability. In addition, the proposed method provides some references for other selection problems.


Sign in / Sign up

Export Citation Format

Share Document