Credit risk, also known as default risk, is to evaluate the possibility of a debt suffering losses due to certain factors. Therefore, the credit evaluation system must be able to not only have the ability to identify defaults, but also be interpretable and provide reasons for defaults.
The construction of the listed companies' credit risk system involves at least three problems. The first problem is selecting a combination of features, as 2m-1 feature combinations exist for m features. Further, different feature combinations from the same sample will have different evaluation results. Therefore, a scientific issue that this article aims to explore and answer is how to build a feature selection model and select a group of features with the greatest discriminative ability. The second problem is determining the feature weights, as the classification results obtained by assigning different weights for the same set of features will differ, or even oppose one another. In a classification model that includes feature weights, the weight setting directly affects the model's classification ability. Therefore, another issue that this article aims to address is how to reasonably weigh the system features. The third problem involves constructing an optimal classification model. Traditional classification models, such as the support vector machines and neural network models, among others, have a theoretical classification threshold of 0.5. As a sample's optimal classification threshold will differ, determining whether its selection is reasonable will directly affect the model's classification performance. Further, the previously mentioned models are "black box" models, and cannot explain the causes of defaults. Consequently, another issue that must be addressed is how to choose an interpretable, highly accurate classification model.
The following solutions for these three problems are proposed in this work. Firstly, the feature system with the largest discrimination ability is selected by the generalized F-Score with the largest discrimination and the smallest Pearson correlation coefficient among the features. Secondly, a set of optimal weights is obtained by constructing a combination weighting model, taking the maximum feature information content and the maximum consistency of weighting results as the objective function. Thirdly, through minimizing the generalized weighted distance between samples and categories, the relative membership degrees of the customer in different categories are obtained by solving the non-linear goal programming, and then the category of the customer is obtained.
An empirical analysis of the listed company samples shows that:(1) 31 features selected from 610 features have the ability to distinguish the defaults and non-defaults and conform to the principle of five Cs of credit, which is a scientific and reasonable feature system; (2) Weights of 31 features are obtained by combination weighting model, and financial factors within enterprises have the greatest impact on credit risk, and working capital turnover ratio having the greatest weight is a key feature affecting the credit risk. The number of supervisory committee meeting, dividend payout ratio, type of audit opinion are all important features affecting the credit rating; (3) Compared with eight models such as logistic regression and support vector machine, etc., it is found that the model proposed in this paper not only has good classification performance, but also is a model balancing classification accuracy and complexity.
[1] Traczynski J. Firm default prediction:A Bayesian model-averaging approach[J]. Journal of Financial and Quantitative Analysis, 2017, 52(3):1211-1245.
[2] Bai Chunguang, Shi Baofeng, Liu Feng, et al. Banking credit worthiness:Evaluating the complex relationships[J]. Omega, 2019, 83:26-38.
[3] 程砚秋, 徐占东. 基于泰尔指数修正的ELECTRE Ⅲ 小企业信用评价模型[J]. 中国管理科学, 2019, 27(10):22-33. Cheng Yanqiu, Xu Zhandong. Credit risk evaluation of small enterprises based on revised ELECTRE Ⅲ by Theil index[J]. Chinese Journal of Management Science, 2019, 27(10):22-33.
[4] 葛兴浪, 刘海龙. 基于动态权重的信用评级[J]. 系统管理学报, 2019, 28(2):285-293. Ge Xinglang, Liu Hailong. Credit rating based on dynamic weight optimization[J]. Journal of Systems & Management, 2019, 28(2):285-293.
[5] Koutanaei F N, Sajedi H, Khanbabaei M. A hybrid data mining model of feature selection algorithms and ensemble learning classifiers for credit scoring[J]. Journal of Retailing and Consumer Services, 2015, 27:11-23.
[6] Chi Guotai, Yu Shanli, Zhou Ying. A novel credit evaluation model based on the maximum discrimination of evaluation results[J]. Emerging Markets Finance and Trade, 2019, 56(11):2543-2562.
[7] Jadhav S, He H, Jenkins K. Information gain directed genetic algorithm wrapper feature selection for credit rating[J]. Applied Soft Computing, 2018, 69:541-553.
[8] Chen Feilong, Li Fengchia. Combination of feature selection approaches with SVM in credit scoring[J]. Expert Systems with Applications, 2010, 37(7):4902-4909.
[9] Wang Hong, Xu Qingsong, Zhou Lifeng. Large unbalanced credit scoring using Lasso-logistic regression ensemble[J]. PloS one, 2015, 10(2):1-20.
[10] 陈晓红, 杨志慧. 基于改进模糊综合评价法的信用评估体系研究-以我国中小上市公司为样本的实证研究[J].中国管理科学, 2015, 23(1):146-153. Chen Xiaohong, Yang Zhihui. The study of credit evaluation system based on improved fuzzy evaluation method-Evidence from small and medium-sized enterprises in China[J]. Chinese Journal of Management Science, 2015, 23(1):146-153.
[11] 谢文静, 鲍新中, 董文妍. B2B电商平台供应链融资风险的影响因素识别与评价[J]. 财会月刊, 2020, 869(1):118-126. Xie Wenjing, Bao Xinzhong, Dong Wenyan. Influencing factors' identification and evaluation of supply chain financing risk in B2B ecommerce platform[J]. Finance and Accounting Monthly, 2020, 869(1):118-126.
[12] Karaaslan A, Özden K Ö. Forecasting turkey's credit ratings with multivariate grey model and grey relational analysis[J]. Journal of Quantitative Economics, 2017, 15(3):583-610.
[13] Chai Nana, Wu Bi, Yang Weiwei, et al. A multicriteria approach for modeling small enterprise credit rating:Evidence from China[J]. Emerging Markets Finance and Trade, 2019, 55(11):2523-2543.
[14] Liu Huchen, You Jianxin, You Xiaoyue, et al. A novel approach for failure mode and effects analysis using combination weighting and fuzzy VIKOR method[J]. Applied Soft Computing, 2015, 28:579-588.
[15] 迟国泰, 李鸿禧, 潘明道. 基于违约鉴别能力组合赋权的小企业信用评级-基于小型工业企业样本数据的实证分析[J]. 管理科学学报, 2018, 21(3):105-126. Chi Guotai, Li Hongxi, Pan Mingdao. Small enterprises credit rating based on default identification ability of combination weighting:Based on an empirical analysis of small industrial enterprises[J]. Journal of Management Sciences in China, 2018, 21(3):105-126.
[16] 孟斌, 杨越, 刁姝杰. 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[J]. 系统工程理论与实践, 2019, 39(2):346-359. Meng Bin, Yang Yue, Diao Shujie. Credit evaluation model of small construction enterprises based on discriminating between two types of customers[J]. Systems Engineering-Theory & Practice, 2019, 39(2):346-359.
[17] 齐佳音, 马君, 肖丽妍, 等. 考虑客户风险修正的客户终生价值建模[J]. 管理工程学报, 2015, 29(2):149-159. Qi Jiayin, Ma Jun, Xiao Liyan, et al. Risk-adjusted customer lifetime value measurement[J]. Journal of Industrial Engineering and Engineering Management, 2015, 29(2):149-159.
[18] Sohn S Y, Kim D H, Yoon J H. Technology credit scoring model with fuzzy logistic regression[J]. Applied Soft Computing, 2016, 43:150-158.
[19] Bequé A, Lessmann S. Extreme learning machines for credit scoring:An empirical evaluation[J]. Expert Systems with Applications, 2017, 86:42-53.
[20] Ganji V R, Mannem S N P. Credit card fraud detection using anti-k nearest neighbor algorithm[J]. International Journal on Computer Science and Engineering, 2012, 4(6):1035-1039.
[21] Zhu Xiaoqian, Li Jianping, Wu Dengsheng, et al. Balancing accuracy, complexity and interpretability in consumer credit decision making:A C-TOPSIS classification approach[J]. Knowledge-Based Systems, 2013, 52:258-267.
[22] Reyes O, Morell C, Ventura S. Effective lazy learning algorithm based on a data gravitation model for multi-label learning[J]. Information Sciences, 2016, 340-341:159-174.
[23] 龚承柱, 李兰兰, 卫振锋, 等. 基于前景理论和隶属度的混合型多属性决策方法[J].中国管理科学, 2014, 22(10):122-128. Gong Chenzhu, Li Yanlan, Wei Zhenfeng, et al. A method for hybrid multiple attribute decision making based on prospect theory and membership[J]. Chinese Journal of Management Science, 2014, 22(10):122-128.
[24] Akita T, Miyata S. The bi-dimensional decomposition of regional inequality based on the weighted coefficient of variation[J]. Letters in Spatial and Resource Sciences, 2010, 3(3):91-100.
[25] 李柏洲, 齐鑫, 徐广玉. 基于组合赋权模型的区域知识创造能力评价研究-31个省市自治区视阈的实证分析[J]. 运筹与管理, 2016, 25(3):178-185. Li Baizhou, Qi Xin, Xu Guangyu. Regional knowledge creation model based on optimal combination weight-The empirical analysis of 31 province scope[J]. Operations Research and Management Science, 2016, 25(3):178-185.
[26] Baghai R P, Becker B. Reputations and credit ratings:Evidence from commercial mortgage-backed securities[J]. Journal of Financial Economics, 2020, 135(2):425-444.
[27] 石宝峰, 刘锋, 王建军, 等. 基于PROMETHEE-Ⅱ 的商户小额贷款信用评级模型及实证[J]. 运筹与管理, 2017, 26(9):137-147. Shi Baofeng, Liu Feng, Wang Jianjun, et al. A credit rating model of microfinance loans for small private business based on PROMETHEE-Ⅱ and its empirical study[J]. Operations Research and Management Science, 2017, 26(9):137-147.
[28] 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 Binqing, Yang Yang, Li Xindan, et al. Research on the credit rating of small and micro enterprises based on fuzzy neural network[J]. Journal of Management Sciences in China, 2016, 19(11):114-126.
[29] 张发明, 李艾珉, 韩媛媛. 基于改进动态组合评价方法的小微企业信用评价研究[J]. 管理学报, 2019, 16(2):286-296. Zhang Faming, Li Aimin, Han Yuanyuan. Study on small and micro businesses credit assessment based on improved dynamic combined evaluation method[J]. Chinese Journal of Management, 2019, 16(2):286-296.
[30] 陈林, 谢彦妩, 李平, 等. 借款陈述文字中的违约信号-基于P2P网络借贷的实证研究[J]. 中国管理科学, 2019, 27(4):37-47. Chen Lin, Xie Yanwu, Li Ping, et al. The signal of default risk from the description-text based on the empirical research of P2P lending[J]. Chinese Journal of Management Science, 2019, 27(4):37-47.
[31] Ninh B P V, Do Thanh T, Hong D V. Financial distress and bankruptcy prediction:An appropriate model for listed firms in Vietnam[J]. Economic Systems, 2018, 42(4):616-624.
[32] 李刚, 张亚京, 王基凡, 等. 基于Wilcoxon检验的小型工业企业债信评级模型及实证[J]. 管理评论, 2019, 31(11):3-19. Li Gang, Zhang Yajing, Wang Jifan, et al. Research on the small industrial business credit rating model and empirical[J]. Management Review, 2019, 31(11):3-19.
[33] Antunes A, Bonfim D, Monteiro N, et al. Forecasting banking crises with dynamic panel probit models[J]. International Journal of Forecasting, 2018, 34(2):249-275.
