Diseño de modelo Credit Scoring para evaluación del riesgo crediticio de la cartera de créditos consumo de la Caja Trujillo: 2011 – 2018

• Autor: Valdiviezo González, Patricia Giovanna
• Formato: tesis de grado
• Fecha de Publicación: 2019
• Institución: Universidad Nacional de Trujillo
• Repositorio: UNITRU-Tesis
• Lenguaje: español
• OAI Identifier: oai:dspace.unitru.edu.pe:20.500.14414/13493
• Enlace del recurso: https://hdl.handle.net/20.500.14414/13493
• Nivel de acceso: acceso abierto
• Materia: Riesgo crediticio; Modelo Credit Scoring; Credito consumo; Regresión logística; Curva ROC

The objective of this research is to design a Credit Scoring model for the credit risk assessment of the consumer loan portfolio of the Trujillo Municipal Savings and Credit Fund. From the data from a consolidated base of 54896 customers with consumer loans granted during the period of 2011 - 2018, a sample of 12740 customers was obtained from whom, depending on the characteristics of the borrower, the credit conditions and the inclusion of variables that reflect the macroeconomic environment, a logistic regression was estimated to obtain the probability of default of the loan applicants. The results showed that they are the variables directly related to the client (Territorial Zone, Client Type Caja Trujillo, Marital Status, Age of the client and Income) as well as the variables related to credit (Type of credit or product, Amount, Interest rate, Active rate in national currency and quota), which formed the estimate of a correctly adjusted Credit Scoring model. The goodness of fit tests to which the model was submitted, showed the predictive power of the model by indicating a global percentage of classification of 94.1% and an acceptable measure of the ROC Curve of an area under the curve of 0.8325 which indicated the high discriminatory power between the classification of good payers and bad payers.