目前商业银行面临的个人信用风险问题极其复杂，如何对个人信用风险进行管理非常重要。个人信用风险建模是其中很关键的一步。利用某商业银行信用卡数据，构建信用评分模型，预测客户的违约概率。通过采用ROSE（random over sampling examples）方法处理类别不均衡的问题，利用Group-Lasso（AUC准则）方法进行变量选择，构建基于Logistic回归的信用评分模型。实证结果表明，该方法对样本数据进行类别不均衡处理的结果比其他模型在判别能力和预测能力上更为有效。采用该方法所构建的模型能够作为客户信用评价决策的有效依据，指导银行及其他金融机构评估顾客个人信用风险，在实际运用中具有良好的可操作性。
In view of the complexity of the customers' credit risk faced by commercial banks at the present, how to manage customers' credit risk is very important. Customers' credit risk modeling is a key step. We use the credit card data of a commercial bank to construct a credit scoring model and predict the default probability. We construct a credit scoring model on the basis of Logistic regression, using the group-Lasso (AUC criterion) method to select variables and using the ROSE (random over sampling examples) method to deal with the unbalanced categories. The results are compared and analyzed, and the new model constructed in this work has certain advantages in discriminating ability and predictive ability. It can play a guiding role for banks and other financial institutions in evaluating customer credit risk and can be used as an effective basis for customer credit evaluation decision. In practice, it also has good operability.
[1] 巴曙松. 巴塞尔新资本协议研究[M]. 北京:中国金融出版社,2003.
[2] 王颖,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.
[3] Stapor K, Smolarczyk T, Fabian P. Heteroscedastic discriminant analysis combined with feature selection for credit scoring[J]. Statistics in Transition New Series, 2016, 17(2):265-280.
[4] Kordas G. Credit scoring using binary quantile regression[C]//Dodge Y. Statistical Data Analysis Based on the L1-Norm and Related Methods. Basel:Birkhäuser Press, 2002:125-137.
[5] Baesens B, Setiono R, Mues C, et al. Using neural network rule extraction and decision tables for credit-risk evaluation[J]. Management Science, 2003, 49(3):312-329.
[6] Marcano-Cedeño A, Marin-De-La-Barcena A, Jimenez-Trillo J, et al. Artificial metaplasticity neural network applied to credit scoring[J]. International Journal of Neural Systems, 2011, 21(4):311-317.
[7] Desai V S, Conway D G, Crook J N, et al. Credit-scoring models in the credit-union environment using neural networks and genetic algorithms[J]. IMA Journal of Management Mathematics, 1997, 8(4):323-346.
[8] Lundy M. Cluster analysis in credit scoring:credit scoring and credit control[M]. New York:Oxford University Press, 1993.
[9] Henley W E, Hand D J. A k-nearest-neighbour classifier for assessing consumer credit risk[J]. Journal of the Royal Statistical Society, 1996, 45(1):77-95.
[10] Khanbabaei M, Alborzi M. The use of genetic algorithm:clustering and feature selection techniques in construction of decision tree models for credit scoring[J]. International Journal of Managing Information Technology, 2013, 5(4):13-32.
[11] Bunn D W. An empirical Bayes procedure for the credit granting decision[J]. Operations Research Letters, 1981, 1(1):10-12.
[12] 刘喜和,郭娜. 我国住房抵押贷款信用风险因素分析[J]. 山东社会科学,2012(3):105-108.
[13] Lee T H, Zhang M. Bias correction and statistical test for developing credit scoring model through logistic regression approach[J]. International Journal of Information Technology & Decision Making, 2003, 2(2):299-311.
[14] De Jongh P J, De Jongh E, Pienaar M, et al. The impact of pre-selected variance inflation factor thresholds on the stability and predictive power of logistic regression models in credit scoring[J]. African Journals OnlineORiON, 2015, 31(1):17-37.
[15] 魏秋萍,张景肖. 基于偏最小二乘方法的信用评分模型[J]. 统计与决策,2012(10):4-6.
[16] 史小康,何晓群. 有偏Logistic回归模型及其在个人信用评级中的应用研究[J]. 数理统计与管理,2015,34(6):1048-1056.
[17] Tibshirani R. Regression shrinkage and selection via the lasso:a retrospective[J]. Journal of the Royal Statistical Society B, 2011, 73(3):273-282.
[18] 张婷婷,景英川. 个人信用评分的Adaptive Lasso-Logistic回归分析[J]. 数学的实践与认识,2016,46(18):92-99.
[19] 张娟,张贝贝. 基于Group-Lasso方法的广义半参数可加信用评分模型应用研究[J]. 数理统计与管理,2016,35(3):517-524.
[20] Meier L, Van de Geer S, Bühlmanm P. The group lasso for logistic regression[J]. Journal of the Royal Statistical Society B, 2008, 70(1):53-71.
[21] Zou H, Hastie T, Tibshirani R. On the "degrees of freedom" of the lasso[J]. The Annals of Statistics, 2007, 35(5):2173-2192.
[22] Yuan M, Lin Y. Model selection and estimation in regression with grouped variables[J]. Journal of the Royal Statistical Society B, 2006, 68(1):49-67.
[23] Tseng P. Convergence of a block coordinate descent method for nondifferentiable minimization[J]. Journal of Optimization Theory and Applications, 2001, 109(3):475-494.
[24] Chawla N V, Japkowicz N, Kotcz A. Editorial:special issue on learning from imbalanced data sets[J]. ACM SIGKDD Explorations Newsletter, 2004, 6(1):1-6.
[25] Chawla N V, Bowyer K W, Hall L O, et al. SMOTE:synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16(1):321-357.
[26] Hastie T, Tibshirani R, Friedman J. The elements of statistical learning[M]. 2nd ed. New York:Springer, 2009.
[27] Menardi G, Torelli N. Training and assessing classification rules with imbalanced data[J]. Data Mining and Knowledge Discovery, 2014, 28(1):92-122.
[28] Lunardon N, Menardi G, Torelli N. ROSE:a package for binary imbalanced learning[J]. The R Journal, 2014, 6(1):79-89.
[29] Mamitsuka H. Selecting features in microarray classification using ROC curves[J]. Pattern Recognition, 2006, 39(12):2393-2404.
