Desenvolvimento de Modelos Credit Scoring com Abordagem de Regrassão de Logística para a Gestão da Inadimplência de uma Instituição de Microcrédito

Abstract

    Credit Scoring models are usually used by financial institutions for measurement of credit risk, so are considered a consolidated methodology during the credit concession process of these institutions. This work examines the possibility of application of Credit Scoring models in a micro credit institution denominated Fundo Rotativo de Ação da Cidadania - Cred Cidadania. For this purpose, a sample relative to accounts of Cred Cidadania database was collected and used to develop two types of Credit Scoring models: the first one relating to credit approval process and another named behavioral scoring whic  includes complementary variables. The statistical technique used in construction of models was the logistic regression. The results show that Credit Scoring models obtained satisfactory performance when used in credit risk analysis for the microcredit institution Cred Cidadania, and as supporting instrument of this process. The results also indicated that application of Credit Scoring models supplies subsidies to the institution, assisting it in the prevention and decreasing of insolvency and in the reduction of its operational costs, two problems that affect the financial sustainability.