基于显著区分两类客户的小型建筑企业信用评价模型研究
|
摘要
本文以中国某区域性商业银行185个小型建筑企业的贷款客户为样本,将熵值法权重、CRITIC法权重和方差齐性检验法权重进行组合,通过构建非线性目标规划函数反推出单一赋权方法的组合系数,构建了显著区分违约和非违约客户的小型建筑企业信用评价模型.通过ROC曲线原理,对不同赋权模型的结果进行违约判别能力的检验.本文的创新与特色:一是通过组合赋权得到的信用得分的组内平方和越小、组间平方和越大、那么违约与非违约客户差异越显著的思路建立非线性目标规划模型,通过目标函数最大反推出单一赋权权重的组合系数,保证组合权重的大小能够显著区分客户的违约状态.解决了信用风险评价中组合权重的大小必须对违约状态有显著鉴别能力的难题,避免了现有研究的信用评分模型由于忽略指标权重区分违约状态的能力、导致出现越是可能违约的客户、信用得分反而越高的不合理现象,开拓了信用风险评价指标赋权的新思路.二是根据违约样本均值偏离全部样本均值程度越大、这个指标区分违约状态能力越强、权重越大的思路对指标进行赋权,通过方差齐性检验F值刻画指标的权重,使指标权重的大小反映指标鉴别违约状态能力的大小,改变了现有研究的指标客观赋权方法与违约状态鉴别能力无关的弊端.实证结果表明,与单一赋权结果相比,组合赋权模型的灵敏度和特异度分别为83.33%和95.95%,对客户的违约判别能力更强.
This paper takes 185 loan customers of small and medium construction enterprises in a Chinese regional commercial bank as the sample. It makes a combination of entropy weight, CRITIC method weight and homogeneity of variance weight, by constructing nonlinear goal programming function, the combination coefficient of single empowerment method is deduced. It establishes a credit evaluation model of small construction enterprises that distinguish between default and non default customers. Through the principle of ROC curve, it tests the default judgment ability of different weighting model results. The contributions of this paper are in two aspects. First, it creates the non-linear goal programming function by minimizing the sum of squares between types, and maximizing the sum of squares within each type.The smaller, the smaller the difference between credit scores for customers of the same type. The larger,the larger the difference in average credit scores between types. The smaller, and the larger, the larger the value for the goal programming function. As such, we maximize the value for the goal programming function, to ensure we maximize the difference in credit scores between good and bad customers and are able to clearly differentiate between good and bad customers. Greater mean deviation of default sample from the whole sample lead to bigger deviation from non-default sample as well, and the indicator can easily distinguish default and non-default sample. According to this rule, we assign a larger weight to the index able to identify default state by F value, different form existing index system, which ignores that ability.The results show that the combination weighting model shows better effect than the single weighting model in the capacity of default judgment. Combination weighting model's sensitivity and specificity are 83.33%and 95.95%, and both better than the single weighting model.
This paper takes 185 loan customers of small and medium construction enterprises in a Chinese regional commercial bank as the sample. It makes a combination of entropy weight, CRITIC method weight and homogeneity of variance weight, by constructing nonlinear goal programming function, the combination coefficient of single empowerment method is deduced. It establishes a credit evaluation model of small construction enterprises that distinguish between default and non default customers. Through the principle of ROC curve, it tests the default judgment ability of different weighting model results. The contributions of this paper are in two aspects. First, it creates the non-linear goal programming function by minimizing the sum of squares between types, and maximizing the sum of squares within each type.The smaller, the smaller the difference between credit scores for customers of the same type. The larger,the larger the difference in average credit scores between types. The smaller, and the larger, the larger the value for the goal programming function. As such, we maximize the value for the goal programming function, to ensure we maximize the difference in credit scores between good and bad customers and are able to clearly differentiate between good and bad customers. Greater mean deviation of default sample from the whole sample lead to bigger deviation from non-default sample as well, and the indicator can easily distinguish default and non-default sample. According to this rule, we assign a larger weight to the index able to identify default state by F value, different form existing index system, which ignores that ability.The results show that the combination weighting model shows better effect than the single weighting model in the capacity of default judgment. Combination weighting model's sensitivity and specificity are 83.33%and 95.95%, and both better than the single weighting model.
引文
[1] Hirth S. Credit rating dynamics and competition[J].Journal of Banking&Finance, 2014, 49:100-112.
[2] Weng X, Lii X, Wu Y, et al. Research on the credit risk management of small and medium-sized enterprises based on supply chain finance[J]. Journal of Finance and Accounting, 2016, 4(5):245-253.
[3] Fernandes G B, Artes R. Spatial dependence in credit risk and its improvement in credit scoring[J]. EuropeanJournal of Operational Research, 2016, 249(2):517-524.
[4]张国政,陈维煌,刘呈辉.基于Logistic模型的商业银行个人消费信贷风险评估研究[J].金融理论与实践,2015, 3:53-57.Zhang G Z, Chen W H, Liu C H. Risk assessment of personal consumption credit of commercial banks based on logistic model[J]. Financial Theory&Practice, 2015, 3:53-57.
[5]迟国泰,张亚京,石宝峰.基于Probit回归的小企业债信评级模型及实证[J]·管理科学学报,2016(6):136-156.Chi G T, Zhang Y J, Shi B F. Credit rating model and empirical study of small enterprises based on probit egression[J]. Journal of Management Sciences in China, 2016(6):136-156.
[6] Wei N, Su Y. Credit risk evaluation of SMEs based on supply chain financing[J]. Management Science and Engineering, 2016, 10(2):51-56.
[7]肖斌卿,杨旸,余哲,等.小微企业信用评级模型及比较研究[J]·系统工程学报,2016, 31(6):798-807.Xiao B Q, Yang Y, Yu Z, et al. Credit rating model and comparative study of small and micro enterprise[J].Journal of Systems Engineering, 2016, 31(6):798-807.
[8]姚潇,余乐安.模糊近似支持向量机模型及其在信用风险评估中的应用[J]·系统工程理论与实践,2012, 32(3):549-554.Yao X, Yu L A. Fuzzy approximate support vector machine model and its application in credit risk assessment[J].Systems Engineering—Theory&Practice, 2012, 32(3):549-554.
[9] Siami M, Gholamian M R, Basiri J. An application of locally linear model tree algorithm with combination of feature selection in credit scoring[J]. International Journal of SysteIms Science, 2013, 23:1-10.
[10] Niklis D, Doumpos M, Zopounidis C. Combining market and accounting-based models for credit scoring using a classification scheme based on support vector machines[J]. Applied Mathematics and Computation, 2014, 234:69-81.
[11] Tsai C F, Hung C. Modeling credit scoring using neural network ensembles[J]. Kybernetes, 2014, 43(7):1114-1123.
[12] Harris T. Credit scoring using the cliustered support vector machine[J]. Expert Systems with Applications, 2015,42(2):741-750.
[13]张奇,胡蓝艺,王压.基于Logit与SVM的银行业信用风险预替模型研究[J].系统工程理论与实践,2015, 35(7):1784-1790.Zhang Q, Hu L Y, Wang J. Research on early warning model of banking credit risk based on Logit and SVM[J].Systems Engineering—Theory&Practice, 2015, 35(7):1784-1790.
[14]肖斌卿,杨旸,李心丹,等·基于模糊神经网络的小微企业信用评级研究[J].管理科学学报,2016, 19(11):114-126.Xiao B Q, Yang Y, Li X D, et al. Research on credit rating of small and micro enterprises based on fuzzy neural network[J]. Journal of Management Sciences in China, 2016, 19(11):114-126.
[15]周勇,谢尚宇,袁嫒.信用违约风险模型中违约概率的统计推断[J].系统工程理论与实践,2008, 28(8):207-214.Zhou Y, Xie S Y, Yuan Y. Statistical inference of default probability in credit default risk model[J]. Systems Engineering—Theory&Practice, 2008, 28(8):207-214.
[16]王颍,聂广礼,石勇.基于信用评分模型的我国商业银行客户违约概率研究[J].管理评论,2012, 24(2):78-87.Wang Y, Nie G L, Shi Y. Research on customer default probability of Chinese Commercial Banks Based on credit scoring model[J]. Management Review, 2012, 24(2):78-87.
[17]曾诗鸿,王芳.基于KMV模型的制造业上市公司信用风险评价研究[J].预测,2013, 32(2):60-69.Zeng S H, Wang F. Research on credit risk evaluation of listed companies in manufacturing industry based on KMV model[J]. Forecasting, 2013, 32(2):60-69.
[18]张朝辉,梁慧,褚美玲.基于P2P模式的C2C信用评价方法研究[J].管理评论,2015, 27(12):50-56.Zhang C H, Liang H, Chu M L. Research on C2C credit evaluation method based on P2P model[,J]. Management Review, 2015, 27(12):50-56.
[19] Zhou X, Jiang W, Shi Y, et al. Credit risk evaluation with kernel-based affine subspace nearest points learning method[J]. Expert Systems with Applications, 2011, 38(4):4272-4279.
[20]徐晓萍,马文杰.非上市中小企业贷款违约率的定量分析——基于判别分析法和决策树模型的分析[J].金融研究,2011,3:111-120.Xu X P, Ma W J. Quantitative analysis of default rate of unlisted small and medium sized enterprises—Based on discriminant analysis and decision tree model[J]. Journal of Financial Research, 2011, 3:111一120.
[21] Meng B, Chi G T. New combined weighting model based on maximizing the difference in evaluation results and its application[J]. Mathematical Problems in Engineering, 2015, Article ID 239634, 9 pages, doi:10.1155/2015/239634.
[22]迟国泰,潘明道,程砚秋.基于综合判别能力的农户小额贷款信用评价模型[J].管理评论,20l5, 27(6):42-57.Chi G T, Pan M D, Cheng Y Q. Credit evaluation model of peasant household micro credit based on coimprehensive discriminant ability[J].Management Review, 2015, 27(6):42-57.
[23]宋媚,张朋柱,范静.基于G2B共享信息中介的异源信息信任形成研究[J].系统工程理论与实践,2015, 35(5):1177-1186.Song M, Zhang P Z, Fan J. Research on trust formation of heterogeneous information based on G2B shared information intermediary[J]. Systems Engineering—Theory&Practice, 2015, 35(5):1177-1186.
[24]庞素琳,王石玉.社会大数据信息下农户信用借款声誉计算模型与应用[J].系统工程理论与实践,2015, 35(4):837-846.Pang S L, Wang S Y. Calculation model and application of farmer credit loan reputation under social big data information[J]. Systems Engineering—Theory&Practice, 2015, 35(4):837-846.
[25]牟刚,袁先智·大数据架构下企业内部信用评级的实证研究[J].系统工程学报,2016, 31(6):808-815.Mu G, Yuan X Z. An empirical study of internal credit rating under the framework of big data[J]. Journal of Systems Engineering, 2016, 31(6):808-815.
[26]迟国泰,王卫.基于科学发展的综合评价理论、方法与应用[M].北京:科学出版社,2009,Chi G T, Wang W. Comprehensive evaluation theory based on scientific development, method and application[J].Beijing:Science Press, 2009.
[27] Meng B, Chi G T. Evaluation index system of green industry based on maximum information content[J]. The Singapore Economic Review, 2018, 63(2):229-248.
[28]工业和信息化部,国家统计局,国家发展和改革委员会,财政部.中小企业划型标准规定[S].工信部联企业[2011]300号,2011-6-18.
[2] Weng X, Lii X, Wu Y, et al. Research on the credit risk management of small and medium-sized enterprises based on supply chain finance[J]. Journal of Finance and Accounting, 2016, 4(5):245-253.
[3] Fernandes G B, Artes R. Spatial dependence in credit risk and its improvement in credit scoring[J]. EuropeanJournal of Operational Research, 2016, 249(2):517-524.
[4]张国政,陈维煌,刘呈辉.基于Logistic模型的商业银行个人消费信贷风险评估研究[J].金融理论与实践,2015, 3:53-57.Zhang G Z, Chen W H, Liu C H. Risk assessment of personal consumption credit of commercial banks based on logistic model[J]. Financial Theory&Practice, 2015, 3:53-57.
[5]迟国泰,张亚京,石宝峰.基于Probit回归的小企业债信评级模型及实证[J]·管理科学学报,2016(6):136-156.Chi G T, Zhang Y J, Shi B F. Credit rating model and empirical study of small enterprises based on probit egression[J]. Journal of Management Sciences in China, 2016(6):136-156.
[6] Wei N, Su Y. Credit risk evaluation of SMEs based on supply chain financing[J]. Management Science and Engineering, 2016, 10(2):51-56.
[7]肖斌卿,杨旸,余哲,等.小微企业信用评级模型及比较研究[J]·系统工程学报,2016, 31(6):798-807.Xiao B Q, Yang Y, Yu Z, et al. Credit rating model and comparative study of small and micro enterprise[J].Journal of Systems Engineering, 2016, 31(6):798-807.
[8]姚潇,余乐安.模糊近似支持向量机模型及其在信用风险评估中的应用[J]·系统工程理论与实践,2012, 32(3):549-554.Yao X, Yu L A. Fuzzy approximate support vector machine model and its application in credit risk assessment[J].Systems Engineering—Theory&Practice, 2012, 32(3):549-554.
[9] Siami M, Gholamian M R, Basiri J. An application of locally linear model tree algorithm with combination of feature selection in credit scoring[J]. International Journal of SysteIms Science, 2013, 23:1-10.
[10] Niklis D, Doumpos M, Zopounidis C. Combining market and accounting-based models for credit scoring using a classification scheme based on support vector machines[J]. Applied Mathematics and Computation, 2014, 234:69-81.
[11] Tsai C F, Hung C. Modeling credit scoring using neural network ensembles[J]. Kybernetes, 2014, 43(7):1114-1123.
[12] Harris T. Credit scoring using the cliustered support vector machine[J]. Expert Systems with Applications, 2015,42(2):741-750.
[13]张奇,胡蓝艺,王压.基于Logit与SVM的银行业信用风险预替模型研究[J].系统工程理论与实践,2015, 35(7):1784-1790.Zhang Q, Hu L Y, Wang J. Research on early warning model of banking credit risk based on Logit and SVM[J].Systems Engineering—Theory&Practice, 2015, 35(7):1784-1790.
[14]肖斌卿,杨旸,李心丹,等·基于模糊神经网络的小微企业信用评级研究[J].管理科学学报,2016, 19(11):114-126.Xiao B Q, Yang Y, Li X D, et al. Research on credit rating of small and micro enterprises based on fuzzy neural network[J]. Journal of Management Sciences in China, 2016, 19(11):114-126.
[15]周勇,谢尚宇,袁嫒.信用违约风险模型中违约概率的统计推断[J].系统工程理论与实践,2008, 28(8):207-214.Zhou Y, Xie S Y, Yuan Y. Statistical inference of default probability in credit default risk model[J]. Systems Engineering—Theory&Practice, 2008, 28(8):207-214.
[16]王颍,聂广礼,石勇.基于信用评分模型的我国商业银行客户违约概率研究[J].管理评论,2012, 24(2):78-87.Wang Y, Nie G L, Shi Y. Research on customer default probability of Chinese Commercial Banks Based on credit scoring model[J]. Management Review, 2012, 24(2):78-87.
[17]曾诗鸿,王芳.基于KMV模型的制造业上市公司信用风险评价研究[J].预测,2013, 32(2):60-69.Zeng S H, Wang F. Research on credit risk evaluation of listed companies in manufacturing industry based on KMV model[J]. Forecasting, 2013, 32(2):60-69.
[18]张朝辉,梁慧,褚美玲.基于P2P模式的C2C信用评价方法研究[J].管理评论,2015, 27(12):50-56.Zhang C H, Liang H, Chu M L. Research on C2C credit evaluation method based on P2P model[,J]. Management Review, 2015, 27(12):50-56.
[19] Zhou X, Jiang W, Shi Y, et al. Credit risk evaluation with kernel-based affine subspace nearest points learning method[J]. Expert Systems with Applications, 2011, 38(4):4272-4279.
[20]徐晓萍,马文杰.非上市中小企业贷款违约率的定量分析——基于判别分析法和决策树模型的分析[J].金融研究,2011,3:111-120.Xu X P, Ma W J. Quantitative analysis of default rate of unlisted small and medium sized enterprises—Based on discriminant analysis and decision tree model[J]. Journal of Financial Research, 2011, 3:111一120.
[21] Meng B, Chi G T. New combined weighting model based on maximizing the difference in evaluation results and its application[J]. Mathematical Problems in Engineering, 2015, Article ID 239634, 9 pages, doi:10.1155/2015/239634.
[22]迟国泰,潘明道,程砚秋.基于综合判别能力的农户小额贷款信用评价模型[J].管理评论,20l5, 27(6):42-57.Chi G T, Pan M D, Cheng Y Q. Credit evaluation model of peasant household micro credit based on coimprehensive discriminant ability[J].Management Review, 2015, 27(6):42-57.
[23]宋媚,张朋柱,范静.基于G2B共享信息中介的异源信息信任形成研究[J].系统工程理论与实践,2015, 35(5):1177-1186.Song M, Zhang P Z, Fan J. Research on trust formation of heterogeneous information based on G2B shared information intermediary[J]. Systems Engineering—Theory&Practice, 2015, 35(5):1177-1186.
[24]庞素琳,王石玉.社会大数据信息下农户信用借款声誉计算模型与应用[J].系统工程理论与实践,2015, 35(4):837-846.Pang S L, Wang S Y. Calculation model and application of farmer credit loan reputation under social big data information[J]. Systems Engineering—Theory&Practice, 2015, 35(4):837-846.
[25]牟刚,袁先智·大数据架构下企业内部信用评级的实证研究[J].系统工程学报,2016, 31(6):808-815.Mu G, Yuan X Z. An empirical study of internal credit rating under the framework of big data[J]. Journal of Systems Engineering, 2016, 31(6):808-815.
[26]迟国泰,王卫.基于科学发展的综合评价理论、方法与应用[M].北京:科学出版社,2009,Chi G T, Wang W. Comprehensive evaluation theory based on scientific development, method and application[J].Beijing:Science Press, 2009.
[27] Meng B, Chi G T. Evaluation index system of green industry based on maximum information content[J]. The Singapore Economic Review, 2018, 63(2):229-248.
[28]工业和信息化部,国家统计局,国家发展和改革委员会,财政部.中小企业划型标准规定[S].工信部联企业[2011]300号,2011-6-18.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |