Application of bayesian additive regression trees in the development of credit scoring models in Brazil

Saved in:
Bibliographic Details
Main Author: Daniel Alves de Brito
Format: Artículo científico
Language:en
Published: Associação Brasileira de Engenharia de Produção 2018
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866592882618531840
author Daniel Alves de Brito
author_facet Daniel Alves de Brito
contents Application of bayesian additive regression trees in the development of credit scoring models in Brazil Daniel Alves de Brito Rinaldo Artes Ingeniería BART Credit Random Forest Machine learning Logistic regression Paper aims This paper presents a comparison of the performances of the Bayesian additive regression trees (BART), Random Forest (RF) and the logistic regression model (LRM) for the development of credit scoring models. Originality It is not usual the use of BART methodology for the analysis of credit scoring data. The database was provided by Serasa-Experian with information regarding direct retail consumer credit operations. The use of credit bureau variables is not usual in academic papers. Research method Several models were adjusted and their performances were compared by using regular methods. Main findings The analysis confirms the superiority of the BART model over the LRM for the analyzed data. RF was superior to LRM only for the balanced sample. The best-adjusted BART model was superior to RF. Implications for theory and practice The paper suggests that the use of BART or RF may bring better results for credit scoring modelling. 2018 artículo científico 0103-6513 https://www.redalyc.org/articulo.oa?id=396754754010 https://www.redalyc.org/journal/3967/396754754010/ https://www.redalyc.org/journal/3967/396754754010/html/ https://www.redalyc.org/journal/3967/396754754010/396754754010.epub https://www.redalyc.org/journal/3967/396754754010/movil 10.1590/0103-6513.20170110 en http://www.redalyc.org/revista.oa?id=3967 Production application/pdf Associação Brasileira de Engenharia de Produção Production (Brasil) Vol.28
format Artículo científico
id redalyc_396754754010
language en
publishDate 2018
publisher Associação Brasileira de Engenharia de Produção
spellingShingle Application of bayesian additive regression trees in the development of credit scoring models in Brazil
Daniel Alves de Brito
Ingeniería
BART
Credit
Random Forest
Machine learning
Logistic regression
Application of bayesian additive regression trees in the development of credit scoring models in Brazil Daniel Alves de Brito Rinaldo Artes Ingeniería BART Credit Random Forest Machine learning Logistic regression Paper aims This paper presents a comparison of the performances of the Bayesian additive regression trees (BART), Random Forest (RF) and the logistic regression model (LRM) for the development of credit scoring models. Originality It is not usual the use of BART methodology for the analysis of credit scoring data. The database was provided by Serasa-Experian with information regarding direct retail consumer credit operations. The use of credit bureau variables is not usual in academic papers. Research method Several models were adjusted and their performances were compared by using regular methods. Main findings The analysis confirms the superiority of the BART model over the LRM for the analyzed data. RF was superior to LRM only for the balanced sample. The best-adjusted BART model was superior to RF. Implications for theory and practice The paper suggests that the use of BART or RF may bring better results for credit scoring modelling. 2018 artículo científico 0103-6513 https://www.redalyc.org/articulo.oa?id=396754754010 https://www.redalyc.org/journal/3967/396754754010/ https://www.redalyc.org/journal/3967/396754754010/html/ https://www.redalyc.org/journal/3967/396754754010/396754754010.epub https://www.redalyc.org/journal/3967/396754754010/movil 10.1590/0103-6513.20170110 en http://www.redalyc.org/revista.oa?id=3967 Production application/pdf Associação Brasileira de Engenharia de Produção Production (Brasil) Vol.28
title Application of bayesian additive regression trees in the development of credit scoring models in Brazil
topic Ingeniería
BART
Credit
Random Forest
Machine learning
Logistic regression
url https://www.redalyc.org/articulo.oa?id=396754754010
https://www.redalyc.org/journal/3967/396754754010/
https://www.redalyc.org/journal/3967/396754754010/html/
https://www.redalyc.org/journal/3967/396754754010/396754754010.epub
https://www.redalyc.org/journal/3967/396754754010/movil