dc.contributorLuiz Alberto Bertucci
dc.contributorHudson Fernandes Amaral
dc.creatorLousanne Cavalcanti Barros
dc.date.accessioned2019-08-12T08:27:35Z
dc.date.accessioned2022-10-03T22:27:32Z
dc.date.available2019-08-12T08:27:35Z
dc.date.available2022-10-03T22:27:32Z
dc.date.created2019-08-12T08:27:35Z
dc.date.issued2014-03-28
dc.identifierhttp://hdl.handle.net/1843/BUBD-ANQRTZ
dc.identifier.urihttp://repositorioslatinoamericanos.uchile.cl/handle/2250/3802770
dc.description.abstractOne of strategy widely used and viewed as a way to strengthen of the credit cooperativism sector has been the incorporation process, which occurred basically in moments under which entities could not present financial conditions to conduct your own business. The management of credit risk in this environment is still very traditional, restricting itself many times in the use of the 5c's of credit - Character, Capacity, Condition, Capital and Collateral. However, to ensure that the credit cooperatives can really serve the financial needs of the community, or specifically of its members, it is important that also provide solid financial structure, with low risk of insolvency. The aim of this study was to develop an model for risk assessment of credit cooperatives, using the cooperative Sicoob Nossacoop as a case study. For the development of this work, considering the steps developing of a credit scoring model, it was used the Logistic Regression model to a database of 5.884 outstanding contracts, until December 2012. To evaluate the quality of fit of the model were used the Hosmer and Lemeshow Test and Chi-square. Both showed satisfactory results. Of the 38 independent variables available and tested in the model, were included 7 beyond to the constant, by forward stepwise method: a) number of years worked; b) receive salary by the cooperative; (c) debit balance; (d) the type of product contracted; e) the amount of parcels; (f) time as customer, and f) salary range. To validate the results analysis were used contingency of forecasts, ROC Curve and Kolmogorov-Smirnov Test. For the ROC Curve, the area under the curve corresponded to a value of 0,7, demonstrating an acceptable power of discrimination. Finally, the Kolmogorov-Smirnov test showed a statistical KS of 29%, ie, the chosen model presented an acceptable discrimination approximate of a good discrimination. Considering the results presented, it is believed that this study has contributed to the discussion about credit risk for the Sicoob Nossacoop. Because of the size of the cooperative, if compared to the large commercial banks, it is possible to realize that the trials based on 5C's of credit, generated positive results. However, in times of expansion, guideline proposal by a strategic planning, it is believed that the use of a model for risk assessment can contribute even more in the decision-making process.
dc.publisherUniversidade Federal de Minas Gerais
dc.publisherUFMG
dc.rightsAcesso Aberto
dc.subjectRisco de Crédito
dc.subjectCredit Scoring
dc.subjectRegressão Logística
dc.titleUm modelo para avaliação de risco em uma cooperativa de crédito: um estudo de caso do Sicoob Nossacoop
dc.typeTese de Doutorado


Este ítem pertenece a la siguiente institución