2002
DOI: 10.1590/s1676-56482002000200006
|View full text |Cite
|
Sign up to set email alerts
|

Reconhecimento de padrões: motodologias estatísticas em crédito ao consumidor

Abstract: A inadimplência é um dos maiores problemas, senão o maior, enfrentado pelas administradoras de cartão de crédito. No estudo deste problema foi criado o conceito de risco, que é essencialmente a probabilidade de não recebimento dos créditos por parte das administradoras de cartões. Alguns autores, Caouette et al. (2000) e Silva (1988) referem-se às técnicas estatísticas multivariadas como ferramentas poderosas na administração do risco envolvido na concessão de crédito pessoal. Este trabalho apresenta a constru… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
2
0
7

Year Published

2011
2011
2021
2021

Publication Types

Select...
6
1

Relationship

0
7

Authors

Journals

citations
Cited by 7 publications
(9 citation statements)
references
References 1 publication
0
2
0
7
Order By: Relevance
“…Consequently, the false positive and false negative rates are 9.3% and 31.58%, respectively, considering a partition of 70% for the training sample. Works that have used Fisher's linear discriminant method, like in consumer credit analysis, Guimarães and Chaves-Neto [12] obtained sensitivity and specificity rates of 92.16% and 92.4%, respectively, bur ignoring the partition training set. In the case of cross-validation following the Monte Carlo procedure, which consisted of performing 100 simulations of the training and test set partition and for each of these simulations, calculated the sensibility and specificity of the LDA classifier, the prediction power of the Fisher Linear discriminant remained consistent, in the sense that the values are close to those of Table 3 (Fig.…”
Section: Resultsmentioning
confidence: 99%
“…Consequently, the false positive and false negative rates are 9.3% and 31.58%, respectively, considering a partition of 70% for the training sample. Works that have used Fisher's linear discriminant method, like in consumer credit analysis, Guimarães and Chaves-Neto [12] obtained sensitivity and specificity rates of 92.16% and 92.4%, respectively, bur ignoring the partition training set. In the case of cross-validation following the Monte Carlo procedure, which consisted of performing 100 simulations of the training and test set partition and for each of these simulations, calculated the sensibility and specificity of the LDA classifier, the prediction power of the Fisher Linear discriminant remained consistent, in the sense that the values are close to those of Table 3 (Fig.…”
Section: Resultsmentioning
confidence: 99%
“…In the same study, analyzing seed color (red and yellow), the accuracy of the model was 68.9%. In consumer credit analysis, Guimarães and Chaves-Neto (2002) applied the discriminant analysis model and obtained sensitivity and specificity rates of 92.16% and 92.4%, respectively. Table 4.…”
Section: Resultsmentioning
confidence: 99%
“…É preciso conhecer os fatores que levam a inadimplência para poder preveni-la. Assim, nessa perspectiva Guimarães e Chaves Neto (2002) trazem a ideia de que a inadimplência é considerada um dos maiores problemas enfrentados pelos administradores, sendo que para alguns é considerada a maior das questões do dia a dia das organizações.…”
Section: Inadimplênciaunclassified
“…A literatura referente à inadimplência demonstra que alguns autores tiveram a iniciativa de analisar essa questão em algumas realidades e setores econômicos dentro do contexto brasileiro. Dentre esses, citam-se os trabalhos de Guimarães e Chaves Neto (2002), Araújo e Carmona (2007), Sehn e Carlini Junior 2007 Barros et al (2015), Campara et al (2016).…”
Section: Evidências Empíricas Brasileirasunclassified
See 1 more Smart Citation