Survival mixture models in behavioral scoring

被引：17

作者：

Alves, Bruno Cardoso ^{[1
]}

Dias, Jose G. ^{[2
]}

机构：

[1] Banco Finantia, Lisbon, Portugal

[2] Inst Univ Lisboa ISCTE IUL, Business Res Unit UNIDE IUL, Lisbon, Portugal

来源：

EXPERT SYSTEMS WITH APPLICATIONS | 2015年 / 42卷 / 08期

关键词：

Credit risk; Behavioral scoring; Survival analysis; Mixture models; BANKRUPTCY PREDICTION; MAXIMUM-LIKELIHOOD; CREDIT; RISK; REGRESSION; ALGORITHM; CLASSIFIERS; PERFORMANCE; EM; IF;

D O I：

10.1016/j.eswa.2014.12.036

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

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. (C) 2014 Elsevier Ltd. All rights reserved.

引用

页码：3902 / 3910

页数：9

共 50 条

[31] A unified Minorization-Maximization approach for estimation of general mixture models
Huang, Xi-fen
Liu, Deng-ge
Zhou, Yun-peng
Zhu, Fei
APPLIED MATHEMATICS-A JOURNAL OF CHINESE UNIVERSITIES SERIES B, 2024, 39 (02) : 343 - 362
[32] Vine copula mixture models and clustering for non-Gaussian data
Sahin, Ozge
Czado, Claudia
ECONOMETRICS AND STATISTICS, 2022, 22 : 136 - 158
[33] Unsupervised Liu-type shrinkage estimators for mixture of regression models
Ghanem, Elsayed
Hatefi, Armin
Usefi, Hamid
STATISTICAL METHODS IN MEDICAL RESEARCH, 2024, 33 (08) : 1376 - 1391
[34] Computationally efficient learning of multivariate t mixture models with missing information
Lin, Tsung-I
Ho, Hsiu J.
Shen, Pao S.
COMPUTATIONAL STATISTICS, 2009, 24 (03) : 375 - 392
[35] Would two-stage scoring models alleviate bank exposure to bad debt?
Abdou, Hussein A.
Mitra, Shatarupa
Fry, John
Elamer, Ahmed A.
EXPERT SYSTEMS WITH APPLICATIONS, 2019, 128 : 1 - 13
[36] Model-based classification using latent Gaussian mixture models
McNicholas, Paul D.
JOURNAL OF STATISTICAL PLANNING AND INFERENCE, 2010, 140 (05) : 1175 - 1181
[37] Sufficient Sample Sizes for Discrete-Time Survival Analysis Mixture Models
Moerbeek, Mirjam
STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL, 2014, 21 (01) : 63 - 67
[38] Robust Mixture of Linear Regression Models
Bashir, Shaheena
Carter, E. M.
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 2012, 41 (18) : 3371 - 3388
[39] UNIFORM CONSISTENCY IN NONPARAMETRIC MIXTURE MODELS
Aragam, Bryon
Yang, Ruiyi
ANNALS OF STATISTICS, 2023, 51 (01) : 362 - 390
[40] Pseudo-stochastic EM for sub-Gaussian α-stable mixture models
Zarei, Shaho
Mohammadpour, Adel
DIGITAL SIGNAL PROCESSING, 2020, 99

← 1 2 3 4 5 →