Survival mixture models in behavioral scoring

This paper introduces a general expert system for behavioral scoring.Survival mixture models can handle unobserved heterogeneity in behavioral scoring.This framework integrates survival, immune fraction, and mixture models.This credit risk model unmixes distinct consumer credit risk patterns.Our application shows that this model accommodates distinct hazard rates. This paper introduces a general framework of survival mixture models (SMMs) that addresses the unobserved heterogeneity of the credit risk of a financial institution's clients. This new behavioral scoring framework contains the specific cases of aggregate and immune fraction models. This general methodology identifies clusters or groups of clients with different risk patterns. The parameters of the model can be explained by independent variables in a regression setting. The application shows the different risk trajectories of clients. Specifically, the time between the first delayed payment and default was best modeled by a three-segment log-normal mixture distribution and a multinomial logit link function. Each segment contains clients with similar risk profiles. The model predicts the most likely risk segment for each new client.

[1]  Thomas W. Lauer,et al.  The bank loan approval decision from multiple perspectives , 2013, Expert Syst. Appl..

[2]  John Banasik,et al.  Not if but when will borrowers default , 1999, J. Oper. Res. Soc..

[3]  Andreas Ziegler,et al.  Consumer credit risk: Individual probability estimates using machine learning , 2013, Expert Syst. Appl..

[4]  Ramayya Krishnan,et al.  Predicting repayment of the credit card debt , 2012, Comput. Oper. Res..

[5]  Tengke Xiong,et al.  Personal bankruptcy prediction by mining credit card data , 2013, Expert Syst. Appl..

[6]  José G. Dias,et al.  The SKM Algorithm: A K-Means Algorithm for Clustering Sequential Data , 2008, IBERAMIA.

[7]  Sebastián Lozano,et al.  A competing risks analysis of the duration of federal target funds rates , 2012, Comput. Oper. Res..

[8]  M. Zmijewski.  METHODOLOGICAL ISSUES RELATED TO THE ESTIMATION OF FINANCIAL DISTRESS PREDICTION MODELS , 1984 .

[9]  William D. Penny,et al.  Comparing Dynamic Causal Models using AIC, BIC and Free Energy , 2012, NeuroImage.

[10]  Maria Stepanova,et al.  Survival Analysis Methods for Personal Loan Data , 2002, Oper. Res..

[11]  T Bellotti,et al.  Credit scoring with macroeconomic variables using survival analysis , 2009, J. Oper. Res. Soc..

[12]  Christophe Mues,et al.  Mixture cure models in credit scoring: If and when borrowers default , 2012, Eur. J. Oper. Res..

[13]  Allen Buxton.  CUREREGR: Stata module to estimate parametric cure regression , 2004 .

[14]  Jonathan Crook,et al.  Modelling profitability using survival combination scores , 2007, Eur. J. Oper. Res..

[15]  S. Upadhyaya,et al.  Comparison of NN and LR classifiers in the context of screening native American elders with diabetes , 2013, Expert Syst. Appl..

[16]  José G. Dias,et al.  An empirical comparison of EM, SEM and MCMC performance for problematic Gaussian mixture likelihoods , 2004, Stat. Comput..

[17]  Brian Mac Namee,et al.  A window of opportunity: Assessing behavioural scoring , 2013, Expert Syst. Appl..

[18]  Stjepan Oreski,et al.  Genetic algorithm-based heuristic for feature selection in credit risk assessment , 2014, Expert Syst. Appl..

[19]  José G. Dias,et al.  A bootstrap-based aggregate classifier for model-based clustering , 2008, Comput. Stat..

[20]  Michel Wedel,et al.  Concomitant variables in finite mixture models , 2002 .

[21]  Gianluca Bontempi,et al.  Learned lessons in credit card fraud detection from a practitioner perspective , 2014, Expert Syst. Appl..

[22]  L. Thomas.  A survey of credit and behavioural scoring: forecasting financial risk of lending to consumers , 2000 .

[23]  Paul C. Lambert,et al.  Modeling of the Cure Fraction in Survival Studies , 2007 .

[24]  Yildiray Yildirim.  Estimating Default Probabilities of CMBS Loans with Clustering and Heavy Censoring , 2008 .

[25]  D.,et al.  Regression Models and Life-Tables , 2022 .

[26]  A. V. Peterson,et al.  On the regression analysis of multivariate failure time data , 1981 .

[27]  Daniel Rösch,et al.  Cure events in default prediction , 2014, Eur. J. Oper. Res..

[28]  David G. Kleinbaum,et al.  Survival Analysis: A Self-Learning Text , 1997 .

[29]  Nizar Bouguila,et al.  Object clustering and recognition using multi-finite mixtures for semantic classes and hierarchy modeling , 2014, Expert Syst. Appl..

[30]  David J. Hand,et al.  Behavioural models of credit card usage , 2003 .

[31]  Jan Beran,et al.  Credit risk modeling based on survival analysis with immunes , 2007 .

[32]  L. Thomas.  Consumer credit models: pricing, profit and portfolios , 2009 .

[33]  Özden Gür Ali,et al.  Dynamic churn prediction framework with more effective use of rare event data: The case of private banking , 2014, Expert Syst. Appl..

[34]  Joaquín Abellán,et al.  Improving experimental studies about ensembles of classifiers for bankruptcy prediction and credit scoring , 2014, Expert Syst. Appl..

[35]  Nebojsa Nikolic,et al.  The application of brute force logistic regression to corporate credit scoring models: Evidence from Serbian financial statements , 2013, Expert Syst. Appl..

[36]  D. Kleinbaum,et al.  Survival Analysis: A Self-Learning Text. , 1996 .

[37]  Jay Magidson,et al.  Latent Class Analysis With Sampling Weights , 2007 .

[38]  William T. Scherer,et al.  Time will tell: behavioural scoring and the dynamics of consumer credit assessment , 2001 .

[39]  E. Kaplan,et al.  Nonparametric Estimation from Incomplete Observations , 1958 .

[40]  Tom Fawcett,et al.  An introduction to ROC analysis , 2006, Pattern Recognit. Lett..

[41]  Niccolò Gordini,et al.  A genetic algorithm approach for SMEs bankruptcy prediction: Empirical evidence from Italy , 2014, Expert Syst. Appl..

[42]  James M. Carson,et al.  An Empirical Examination of Sample Selection Methods in the Context of Life Insurer Financial Distress , 2003 .

[43]  Madhur Malik,et al.  Modeling Credit Risk of Portfolio of Consumer Loans , 2007 .

[44]  Geoffrey J. McLachlan,et al.  Finite Mixture Models , 2019, Annual Review of Statistics and Its Application.

[45]  George B. Macready,et al.  Concomitant-Variable Latent-Class Models , 1988 .

[46]  José G. Dias,et al.  Mining categorical sequences from data using a hybrid clustering method , 2014, Eur. J. Oper. Res..

[47]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[48]  D. Cox,et al.  A General Definition of Residuals , 1968 .

[49]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[50]  H NOH,et al.  Prognostic personal credit risk model considering censored information , 2005, Expert Syst. Appl..

[51]  Marijana Zekic-Susac,et al.  Comparison procedure of predicting the time to default in behavioural scoring , 2009, Expert Syst. Appl..

[52]  Nicole Martin,et al.  Assessing scorecard performance: A literature review and classification , 2013, Expert Syst. Appl..