基于混合生存分析的动态信用评分方法

王钊, 蒋翠清, 丁勇

系统工程理论与实践 ›› 2021, Vol. 41 ›› Issue (2) : 389-399.

PDF(730 KB)
PDF(730 KB)
系统工程理论与实践 ›› 2021, Vol. 41 ›› Issue (2) : 389-399. DOI: 10.12011/SETP2020-1095
论文

基于混合生存分析的动态信用评分方法

    王钊1, 蒋翠清1,2, 丁勇1
作者信息 +

Dynamic credit scoring method based on mixture survival analysis

    WANG Zhao1, JIANG Cuiqing1,2, DING Yong1
Author information +
文章历史 +

摘要

当今国内外经济形势复杂多变,不确定因素增多,金融市场中信用风险的动态性显著增强,动态信用风险评价成为金融机构迫切需要解决的问题.为此,本文提出了一种基于混合生存分析的动态信用评分方法.该方法由三部分组成:首先,构建基于混合生存分析的动态信用评分模型,包括违约状态判别模型和违约时间估计模型,用于预测评价对象“是否违约”以及“何时违约”;其次,利用面板数据构建多重生存状态向量,用于表征信用特征和生存时间的动态关联;最后,基于生成的多重生存状态向量,利用期望最大算法迭代估计模型参数.实验研究表明该方法的预测效果显著优于基于单分类、基于集成学习以及基于生存分析的信用评分方法.

Abstract

Currently, the domestic and international economic situation has become more complex and changeable, with increasing uncertainties, and the dynamics of credit risk in the financial market has been significantly enhanced. Dynamic credit risk evaluation has become an urgent problem that financial institutions need to solve. To this end, this paper proposes a mixture survival analysis-based dynamic credit scoring method, which consists of three parts. First, constructing mixture survival analysis-based dynamic credit scoring model, including default status discrimination model and default time estimation method, to predict “whether default” and “when to default” for evaluation objects. Then, generating multiple surviv-al status vectors using panel data to characterize the dynamic correlations between credit features and survival time. Finally, based on generated multiple survival status vectors, iteratively estimating model parameters using the expectation maximum algorithm. Experimental research shows that the predictive performance of the proposed method is significantly superior to the single classification-based, ensemble learning-based, and survival analysis-based credit scoring methods.

关键词

信用风险 / 动态评价 / 混合生存分析 / 违约状态 / 违约时间

Key words

credit risk / dynamic evaluation / mixture survival analysis / default status / default time

引用本文

导出引用
王钊 , 蒋翠清 , 丁勇. 基于混合生存分析的动态信用评分方法. 系统工程理论与实践, 2021, 41(2): 389-399 https://doi.org/10.12011/SETP2020-1095
WANG Zhao , JIANG Cuiqing , DING Yong. Dynamic credit scoring method based on mixture survival analysis. Systems Engineering - Theory & Practice, 2021, 41(2): 389-399 https://doi.org/10.12011/SETP2020-1095
中图分类号: F830.5   

参考文献

[1] 余乐安, 张有德. 基于关联规则赋权特征选择集成的信用分类研究[J]. 系统工程理论与实践, 2020, 40(2):366-372.Yu L A, Zhang Y D. Weight-selected attribute bagging based on association rules for credit dataset classification[J]. Systems Engineering-Theory & Practice, 2020, 40(2):366-372.
[2] 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 F M, Li A M, Han Y Y. 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.
[3] Djeundje V B, Crook J. Dynamic survival models with varying coefficients for credit risks[J]. European Journal of Operational Research, 2019, 275(1):319-333.
[4] Alves B C, Dias J G. Survival mixture models in behavioral scoring[J]. Expert Systems with Applications, 2015, 42(8):3902-3910.
[5] Gupta J, Wilson N, Gregoriou A, et al. The value of operating cash flow in modelling credit risk for SMEs[J]. Applied Financial Economics, 2014, 24(9):649-660.
[6] 方匡南, 赵梦峦. 基于多源数据融合的个人信用评分研究[J]. 统计研究, 2018, 35(12):92-101.Fang K N, Zhao M L. A study on credit scoring based on multi-source data integration[J]. Statistical Research, 2018, 35(12):92-101.
[7] Ma L, Zhao X, Zhou Z, et al. A new aspect on P2P online lending default prediction using meta-level phone usage data in China[J]. Decision Support Systems, 2018, 111:60-71.
[8] Ge R, Feng J, Gu B, et al. Predicting and deterring default with social media information in peer-to-peer lending[J]. Journal of Management Information Systems, 2017, 34(2):401-424.
[9] Chen X, Zhou L, Wan D. Group social capital and lending outcomes in the financial credit market:An empirical study of online peer-to-peer lending[J]. Electronic Commerce Research and Applications, 2016, 15:1-13.
[10] Wang Z, Jiang C, Zhao H, et al. Mining semantic soft factors for credit risk evaluation in peer-to-peer lending[J]. Journal of Management Information Systems, 2020, 37(1):282-308.
[11] 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. Comparative study of credit rating models for small and micro enterprises[J]. Journal of Systems Engineering, 2016, 31(6):798-807.
[12] Baesens B, Van Gestel T, Viaene S, et al. Benchmarking state-of-the-art classification algorithms for credit scoring[J]. Journal of the Operational Research Society, 2003, 54(6):627-635.
[13] Louzada F, Ara A, Fernandes G B. Classification methods applied to credit scoring:Systematic review and overall comparison[J]. Surveys in Operations Research and Management Science, 2016, 21(2):117-134.
[14] 张卫国, 卢媛媛, 刘勇军. 基于非均衡模糊近似支持向量机的P2P网贷借款人信用风险评估及应用[J]. 系统工程理论与实践, 2018, 38(10):2466-2478.Zhang W G, Lu Y Y, Liu Y J. The borrowers' credit risk assessment in P2P platform based on fuzzy proximal support vector machine and its application[J]. Systems Engineering-Theory & Practice, 2018, 38(10):2466-2478.
[15] 郭海湘, 顾明湘, 李诒靖, 等. 基于差分演化的自适应集成学习算法在不均衡数据分类中的应用[J]. 系统工程理论与实践, 2018, 38(5):1284-1299.Guo H X, Gu M Y, Li Y J, et al. An adaptive multiple classifier system based on differential evolution and its application in imbalanced data classification[J]. Systems Engineering-Theory & Practice, 2018, 38(5):1284-1299.
[16] Narain B. Survival analysis and the credit granting decision[M]//Credit Scoring and Credit Control. Oxford:Oxford University Press, Oxford, 1992:109-121.
[17] Tong E N, Mues C, Thomas L C. Mixture cure models in credit scoring:If and when borrowers default[J]. European Journal of Operational Research, 2012, 218(1):132-139.
[18] Luo S, Kong X, Nie T. Spline based survival model for credit risk modeling[J]. European Journal of Operational Research, 2016, 253(3):869-879.
[19] Djeundje V B, Crook J. Identifying hidden patterns in credit risk survival data using generalised additive models[J]. European Journal of Operational Research, 2019, 277(1):366-376.
[20] Dirick L, Claeskens G, Baesens B. Time to default in credit scoring using survival analysis:A benchmark study[J]. Journal of the Operational Research Society, 2017, 68(6):652-665.
[21] Frydman H, Matuszyk A. Random survival forest for competing credit risks[J]. Journal of the Operational Research Society, 2020:1-11. https://doi.org/10.1080/01605682.2020.1759385.
[22] Jiang C, Wang Z, Zhao H. A prediction-driven mixture cure model and its application in credit scoring[J]. European Journal of Operational Research, 2019, 277(1):20-31.
[23] Zhang N, Yang Q, Kelleher A, et al. A new mixture cure model under competing risks to score online consumer loans[J]. Quantitative Finance, 2019, 19(7):1243-1253.
[24] 韩璐, 韩立岩. 正交支持向量机及其在信用评分中的应用[J]. 管理工程学报, 2017, 31(2):128-136.Han L, Han L Y. Orthogonal support vector machine and its application in credit scoring[J]. Journal of Industrial Engineering/Engineering Management, 2017, 31(2):128-136.
[25] Abellán J, Castellano J G. A comparative study on base classifiers in ensemble methods for credit scoring[J]. Expert Systems with Applications, 2017, 73:1-10.
[26] 方匡南, 陈子岚. 基于半监督广义可加Logistic回归的信用评分方法[J]. 系统工程理论与实践, 2020, 40(2):392-402.Fang K N, Chen Z L. Credit scoring based on semi-supervised generalized additive logistic regression[J]. Systems Engineering-Theory & Practice, 2020, 40(2):392-402.
[27] Emekter R, Tu Y, Jirasakuldech B, et al. Evaluating credit risk and loan performance in online Peer-to-Peer (P2P) lending[J]. Applied Economics, 2015, 47(1):54-70.
[28] 蒋辉, 马超群, 许旭庆, 等. 仿EM的多变量缺失数据填补算法及其在信用评估中的应用[J]. 中国管理科学, 2019, 27(3):11-19.Jiang H, Ma C Q, Xu X Q, et al. An EM-similar imputation algorithm for multivariable data missing and its application in credit scoring[J]. Chinese Journal of Management Science, 2019, 27(3):11-19.
[29] Yoon Y, Li Y, Feng Y. Factors affecting platform default risk in online peer-to-peer (P2P) lending business:An empirical study using Chinese online P2P platform data[J]. Electronic Commerce Research, 2019, 19(1):131-158.
[30] Lessmann S, Baesens B, Seow H V, et al. Benchmarking state-of-the-art classification algorithms for credit scoring:An update of research[J]. European Journal of Operational Research, 2015, 247(1):124-136.
[31] Hand D J. Measuring classifier performance:A coherent alternative to the area under the ROC curve[J]. Machine Learning, 2009, 77(1):103-123.

基金

国家自然科学基金重点项目(71731005)
PDF(730 KB)

Accesses

Citation

Detail

段落导航
相关文章

/