Feature selection in credit risk modeling: an international evidence

Abstract

This paper aims to discover a suitable combination of contemporary feature selection techniques and robust prediction classifiers. As such, to examine the impact of the feature selection method on classifier performance, we use two Chinese and three other real-world credit scoring datasets. The utilized feature selection methods are the least absolute shrinkage and selection operator (LASSO), multivariate adaptive regression splines (MARS). In contrast, the examined classifiers are the classification and regression trees (CART), logistic regression (LR), artificial neural network (ANN), and support vector machines (SVM). Empirical findings confirm that LASSO's feature selection method, followed by robust classifier SVM, demonstrates remarkable improvement and outperforms other competitive classifiers. Moreover, ANN also offers improved accuracy with feature selection methods; LR only can improve classification efficiency through performing feature selection via LASSO. Nonetheless, CART does not provide any indication of improvement in any combination. The proposed credit scoring modeling strategy may use to develop policy, progressive ideas, operational guidelines for effective credit risk management of lending, and other financial institutions. The finding of this study has practical value, as to date, there is no consensus about the combination of feature selection method and prediction classifiers.

Effect of variable selection strategy on the predictive models for adverse pregnancy outcomes of pre-eclampsia: A retrospective study

“…With the controlled penalizing coefficient, LASSO facilitates the desirable variables. [20][21][22] As the number of variables selected by the FSLR method was determined automatically, the number of variables selected by the other four strategies was fixed based on the determination made by FSLR. This approach was adopted to facilitate a comparative analysis of the efficiency exhibited by the following models.…”

Identification of Most Preferable Machine Learning Technique for Prediction of Bank Loan Defaulters

“…One of the key benefits of utilizing a Support Vector Machine (SVM) is the ability to procure an optimal hyperplane that effectively divides differing classes through the maximization of the margin between them. It transforms the input data into a high-dimensional feature space and classifies instances based on their position relative to the hyperplane (17) . By using different kernel functions it can handle both linear and non-linear classification problems.…”

Artificial Bee Colony for Logic Mining in Credit Scoring

“…As for the financial statement, issuers can verify the credit ranking of the client based on the default record history and decide to issue the credit [1]. These two categories have strong relationships in predicting the default and choosing new applicants because if they only rely on personal information to issue new credit, this poses a high risk to the issuers since financial information is also essential to be considered.…”

“…Thus, a series of artificial intelligence methods are often used to classify the credit outcome. Zhou et al [1] proposed credit risk modeling by proposing a hybrid support vector machine (SVM). The proposed SVM demonstrates remarkable improvement and outperforms another competitive classifier.…”