A comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysis

Detalhes bibliográficos
Autor(a) principal: Yim, Juliana
Data de Publicação: 2009
Outros Autores: Mitchell, Heather
Tipo de documento: Artigo
Idioma: por
Título da fonte: Nova Economia (Online)
Texto Completo: https://revistas.face.ufmg.br/index.php/novaeconomia/article/view/445
Resumo: This paper looks at the ability of a relatively new technique, hybrid ANN’s, to predict corporate distress in Brazil. These models are compared with traditional statistical techniques and conventional ANN models. The results suggest that hybrid neural networks outperform all other models in predicting firms in financial distress one year prior to the event. This suggests that for researchers, policymakers and others interested in early warning systems, hybrid networks may be a useful tool for predicting firm failure.
id UFMG-4_5a76ee50e89df79bcce4e9b76232a988
oai_identifier_str oai:ojs.pkp.sfu.ca:article/445
network_acronym_str UFMG-4
network_name_str Nova Economia (Online)
repository_id_str
spelling A comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysisA comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysishybrid neural networkscorporate failures.redes neurais híbridasfalência de empresas.This paper looks at the ability of a relatively new technique, hybrid ANN’s, to predict corporate distress in Brazil. These models are compared with traditional statistical techniques and conventional ANN models. The results suggest that hybrid neural networks outperform all other models in predicting firms in financial distress one year prior to the event. This suggests that for researchers, policymakers and others interested in early warning systems, hybrid networks may be a useful tool for predicting firm failure.O presente artigo analisa o desempenho das redes neurais híbridas para prever falência de empresas no Brasil. Esta nova técnica foi comparada com modelos estatísticos tradicionais. Os resultados sugerem que as redes neurais híbridas são superiores as técnicas estatísticas um ano antes do evento. Isto sugere que para pesquisadores, políticos e outros interessados em “early warning systems”, redes neurais híbridas podem ser uma poderosa alternativa para prever falência de empresas.Departamento de Ciências Econômicas da UFMG2009-06-02info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://revistas.face.ufmg.br/index.php/novaeconomia/article/view/445Nova Economia; Vol. 15 No. 1 (2005)Nova Economia; v. 15 n. 1 (2005)1980-53810103-6351reponame:Nova Economia (Online)instname:Universidade Federal de Minas Gerais (UFMG)instacron:UFMGporhttps://revistas.face.ufmg.br/index.php/novaeconomia/article/view/445/442Yim, JulianaMitchell, Heatherinfo:eu-repo/semantics/openAccess2020-08-11T04:26:02Zoai:ojs.pkp.sfu.ca:article/445Revistahttps://revistas.face.ufmg.br/index.php/novaeconomiaPUBhttps://revistas.face.ufmg.br/index.php/novaeconomia/oai||ne@face.ufmg.br1980-53810103-6351opendoar:2020-08-11T04:26:02Nova Economia (Online) - Universidade Federal de Minas Gerais (UFMG)false
dc.title.none.fl_str_mv A comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysis
A comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysis
title A comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysis
spellingShingle A comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysis
Yim, Juliana
hybrid neural networks
corporate failures.
redes neurais híbridas
falência de empresas.
title_short A comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysis
title_full A comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysis
title_fullStr A comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysis
title_full_unstemmed A comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysis
title_sort A comparison of corporate distress prediction models in Brazil: hybrid neural networks, logit models and discriminant analysis
author Yim, Juliana
author_facet Yim, Juliana
Mitchell, Heather
author_role author
author2 Mitchell, Heather
author2_role author
dc.contributor.author.fl_str_mv Yim, Juliana
Mitchell, Heather
dc.subject.por.fl_str_mv hybrid neural networks
corporate failures.
redes neurais híbridas
falência de empresas.
topic hybrid neural networks
corporate failures.
redes neurais híbridas
falência de empresas.
description This paper looks at the ability of a relatively new technique, hybrid ANN’s, to predict corporate distress in Brazil. These models are compared with traditional statistical techniques and conventional ANN models. The results suggest that hybrid neural networks outperform all other models in predicting firms in financial distress one year prior to the event. This suggests that for researchers, policymakers and others interested in early warning systems, hybrid networks may be a useful tool for predicting firm failure.
publishDate 2009
dc.date.none.fl_str_mv 2009-06-02
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv https://revistas.face.ufmg.br/index.php/novaeconomia/article/view/445
url https://revistas.face.ufmg.br/index.php/novaeconomia/article/view/445
dc.language.iso.fl_str_mv por
language por
dc.relation.none.fl_str_mv https://revistas.face.ufmg.br/index.php/novaeconomia/article/view/445/442
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Departamento de Ciências Econômicas da UFMG
publisher.none.fl_str_mv Departamento de Ciências Econômicas da UFMG
dc.source.none.fl_str_mv Nova Economia; Vol. 15 No. 1 (2005)
Nova Economia; v. 15 n. 1 (2005)
1980-5381
0103-6351
reponame:Nova Economia (Online)
instname:Universidade Federal de Minas Gerais (UFMG)
instacron:UFMG
instname_str Universidade Federal de Minas Gerais (UFMG)
instacron_str UFMG
institution UFMG
reponame_str Nova Economia (Online)
collection Nova Economia (Online)
repository.name.fl_str_mv Nova Economia (Online) - Universidade Federal de Minas Gerais (UFMG)
repository.mail.fl_str_mv ||ne@face.ufmg.br
_version_ 1799711057315364864

Registros relacionados