Credit rating system for small businesses using the K-S test to select an indicator system

Purpose The purpose of this paper is to propose a system with the highest discriminatory power by selecting an indicator system based on the K–S test according to the unique circumstances of small enterprises. Design/methodology/approach The proposed method relies on calculating the K–S test statistical magnitude of D iteratively to reach a system with the maximum discriminatory power. Findings The empirical results, demonstrated using 3,045 small businesses from a Chinese bank, show that credit rating system should focus on the indicator system’s discriminatory power rather than a single indicator’s discriminatory power, because the interaction between indicators affects the discriminatory power of the system. Practical implications The proposed method creates a credit rating system with the highest discriminatory power, rather than its indicators, which is a more reasonable and novel approach to credit rating. Originality/value The approach is unique because the final system will have high discriminatory power and has excellent potential for decision support. The authors believe that this contribution is theoretically and practically relevant because credit rating for small business is especially difficult and complicated.

[1]  Allen N. Berger,et al.  Small Business Credit Scoring and Credit Availability* , 2007 .

[2]  Constantin Zopounidis,et al.  Combining Market and Accounting-Based Models for Credit Scoring Using a Classification Scheme Based on Support Vector Machines , 2012, Appl. Math. Comput..

[3]  E. Altman,et al.  Modelling Credit Risk for SMEs: Evidence from the U.S. Market , 2007 .

[4]  Peter Winker,et al.  Optimization heuristics for determining internal rating grading scales , 2010, Comput. Stat. Data Anal..

[5]  F. Ferreira,et al.  Assessing credit risk of mortgage lending using MACBETH: a methodological framework , 2014 .

[6]  Marcus Wunsch,et al.  Moment Estimators for Autocorrelated Time Series and Their Application to Default Correlations , 2017 .

[7]  Mohammad Zoynul Abedin,et al.  Credit default prediction modeling: an application of support vector machine , 2017, Risk Management.

[8]  Sebastián Maldonado,et al.  Cost-based feature selection for Support Vector Machines: An application in credit scoring , 2017, Eur. J. Oper. Res..

[9]  Shu Ling Lin,et al.  A new two-stage hybrid approach of credit risk in banking industry , 2009, Expert Syst. Appl..

[10]  Neil J. Salkind.  BUNDLE: Salkind: Excel Statistics 3e + Salkind: Statistics for People Who (Think They) Hate Statistics (Excel 2010 Edition) 3e + Salkind: Statistics ... Electronic Version (Excel 2010 Edition) 3e , 2015 .

[11]  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..

[12]  E. Altman,et al.  Modeling Credit Risk for Smes: Evidence from the Us Market , 2005 .

[13]  Guotai Chi,et al.  Credit Default Prediction Using a Support Vector Machine and a Probabilistic Neural Network , 2018 .

[14]  Kenneth Carling,et al.  Corporate credit risk modeling and the macroeconomy , 2007 .

[15]  Guotai Chi,et al.  A credit rating model based on a customer number bell-shaped distribution , 2018 .

[16]  Guotai Chi,et al.  A Model for Recognizing Key Factors and Applications Thereof to Engineering , 2014 .

[17]  Longquan Yong,et al.  Low-Rank Representation for Incomplete Data , 2014 .

[18]  P. Winker,et al.  Threshold accepting for credit risk assessment and validation , 2015 .

[19]  Sandra Paterlini,et al.  The optimal structure of PD buckets , 2008 .

[20]  Terry Harris,et al.  Credit scoring using the clustered support vector machine , 2015, Expert Syst. Appl..

[21]  Chih-Fong Tsai,et al.  Modeling credit scoring using neural network ensembles , 2014, Kybernetes.

[22]  Jie Sun,et al.  Combining B&B-based hybrid feature selection and the imbalance-oriented multiple-classifier ensemble for imbalanced credit risk assessment , 2015 .