Feature Selection in a Credit Scoring Model
This paper proposes different classification algorithms—logistic regression, support vector machine, K-nearest neighbors, and random forest—in order to identify which candidates are likely to default for a credit scoring model. Three different feature selection methods are used in order to mitigate...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-03-01
|
Series: | Mathematics |
Subjects: | |
Online Access: | https://www.mdpi.com/2227-7390/9/7/746 |
id |
doaj-e8199b5b3e6f4b9b97a13c1875d5fcf2 |
---|---|
record_format |
Article |
spelling |
doaj-e8199b5b3e6f4b9b97a13c1875d5fcf22021-03-31T23:03:30ZengMDPI AGMathematics2227-73902021-03-01974674610.3390/math9070746Feature Selection in a Credit Scoring ModelJuan Laborda0Seyong Ryoo1Department of Business Administration, University Carlos III, 28903 Madrid, SpainLeuven Statistics Research Centre, KU Leuven, 3000 Leuven, BelgiumThis paper proposes different classification algorithms—logistic regression, support vector machine, K-nearest neighbors, and random forest—in order to identify which candidates are likely to default for a credit scoring model. Three different feature selection methods are used in order to mitigate the overfitting in the curse of dimensionality of these classification algorithms: one filter method (Chi-squared test and correlation coefficients) and two wrapper methods (forward stepwise selection and backward stepwise selection). The performances of these three methods are discussed using two measures, the mean absolute error and the number of selected features. The methodology is applied for a valuable database of Taiwan. The results suggest that forward stepwise selection yields superior performance in each one of the classification algorithms used. The conclusions obtained are related to those in the literature, and their managerial implications are analyzed.https://www.mdpi.com/2227-7390/9/7/746operational research in bankingmachine learningcredit scoringclassification algorithmsfeature selection methods |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Juan Laborda Seyong Ryoo |
spellingShingle |
Juan Laborda Seyong Ryoo Feature Selection in a Credit Scoring Model Mathematics operational research in banking machine learning credit scoring classification algorithms feature selection methods |
author_facet |
Juan Laborda Seyong Ryoo |
author_sort |
Juan Laborda |
title |
Feature Selection in a Credit Scoring Model |
title_short |
Feature Selection in a Credit Scoring Model |
title_full |
Feature Selection in a Credit Scoring Model |
title_fullStr |
Feature Selection in a Credit Scoring Model |
title_full_unstemmed |
Feature Selection in a Credit Scoring Model |
title_sort |
feature selection in a credit scoring model |
publisher |
MDPI AG |
series |
Mathematics |
issn |
2227-7390 |
publishDate |
2021-03-01 |
description |
This paper proposes different classification algorithms—logistic regression, support vector machine, K-nearest neighbors, and random forest—in order to identify which candidates are likely to default for a credit scoring model. Three different feature selection methods are used in order to mitigate the overfitting in the curse of dimensionality of these classification algorithms: one filter method (Chi-squared test and correlation coefficients) and two wrapper methods (forward stepwise selection and backward stepwise selection). The performances of these three methods are discussed using two measures, the mean absolute error and the number of selected features. The methodology is applied for a valuable database of Taiwan. The results suggest that forward stepwise selection yields superior performance in each one of the classification algorithms used. The conclusions obtained are related to those in the literature, and their managerial implications are analyzed. |
topic |
operational research in banking machine learning credit scoring classification algorithms feature selection methods |
url |
https://www.mdpi.com/2227-7390/9/7/746 |
work_keys_str_mv |
AT juanlaborda featureselectioninacreditscoringmodel AT seyongryoo featureselectioninacreditscoringmodel |
_version_ |
1724177095833157632 |