Credit rating is aimed at measuring the possibility of default,so credit rating system must be able to identify default risk,and separate default customers and non-default customers.This paper through thinking of non-default enterprises close to positive ideal point and default enterprises close to negative ideal point,multiobjective programming model is constructed to solve the optimal combination weights.Through the J-T nonparametric test,the rationality of the credit evaluation model is verified.And this paper does empirical analysis about 1 814 loan customers in a regional commercial bank of China.The innovations and characters in this paper:Firstly,minimizing the distance of non-default enterprises' weighted data to positive ideal point as the first objective function,minimizing the distance of default enterprises' weighted data to negative ideal point as the second objective function,multi-objective programming model is constructed to solve the optimal combination weights.The optimal combination weights satisfy the goal "the higher scores of non-default enterprises,the lower scores of default enterprises" and ensure that credit rating model maximizes the distance of non-default enterprises and default enterprises.It changes the disadvantage that combination weights is contrary to credit rating purpose in existing research,and changes the disadvantage that a lot of the score overlap between non-default and default enterprises and the ability of distinguish between non-default and default enterprises is low in existing research.Secondly,this paper use the J-T nonparametric test to verify the rationality of the credit evaluation model,aimed to ensure that the scores of non-default enterprises is higher significantly than the scores of default enterprises,avoids the irrational phenomenon in existing research that the scores of default enterprises is higher.Thirdly,Compared with traditional combination weighting models in existing research,the default identification ability of the combination weighting model in this research is higher than two kinds of combination weighting models in existing research,one is the combination weighting model based on maximum variance,the other is the combination weighting model based on minimum deviation.%信用评级是衡量债务违约的可能性,因此评级体系要有违约风险识别能力,能够将违约客户和非违约客户显著地区分开.通过逼近理想点的思路,构建多目标规划模型求解最优的组合权重,并对中国某区域性商业银行1 814笔小型工业企业贷款进行实证分析.本文的创新与特色一是以非违约企业的数据到正理想点的距离代数和最小为第一个目标函数,以违约企业的数据到负理想点的距离代数和最小为第二个目标函数,构建多目标非线性规划模型进行组合赋权,在满足了“非违约企业的评价得分越高、违约企业的评价得分越低”要求的目标下得到最优的组合赋权的权重系数,使赋权结果保证了评级模型能够将违约企业与非违约企业最大地区分开.改变了现有研究的组合赋权脱离评价目的的弊端,改变了现有研究中违约与非违约企业的评价得分存在大量重叠、对两类企业的区分能力低的弊端.二是通过检验“违约企业的信用得分是否显著小于非违约企业的信用得分”的J-T非参数检验,验证信用评价模型的合理性.改变现有研究忽略对信用评价模型的合理性进行验证的弊端.三是经过实证,发现本研究建立的组合赋权模型的违约鉴别能力(Z =5.546)要高于现有研究的两种常用组合赋权模型、即基于方差最大的组合赋权(Z =4.298)和基于偏差最小的组合赋权(Z=5.182).
展开▼
