Maestría en Administración Financiera · 2017
Modelo Scoring para el otorgamiento de crédito de las Pymes
Scoring models since the 1990s are widely used as methods of predicting the probability of payment of individuals and companies when they access the financial system -- There are several methods for the development of these models, being the most used over time econometric, mathematical and nowadays artificial intelligence; which has made it possible to automate in many financial institutions worldwide credit studies based on the most relevant variables that determine good behavior -- This project is presented as a way of understanding through the construction of a scoring model, which are the most common possible causes for an SME to pay the obligation or not -- This allows to predict the behavior of payment and in turn to comply with the risk policies defined in the risk management system (SARC); which is regulated in Colombia by the Financial Superintendence and based on the frame of reference of Basel Accords. All of this leads to a balance between the pursuit of utility maximization of a financial institution, which is mostly through the placement of resources, and the level of risk that one wishes to assume
Texto completo 28 páginas con texto
Leer la tesis completa Ficha en el repositorio
Contenido
- 1. Introducciónp. 4
- 2. Marco conceptualp. 5
- 2.1 Contexto de las pymes en Colombiap. 5
- 2.2 Riesgo de crédito y cómo se midep. 7
- 2.3 Modelos Scoring, qué son y para qué sirvenp. 8
- 3. Método de soluciónp. 9
- 3.1 Metodología y variables seleccionadasp. 13
- 4. Análisis de resultadosp. 19
- 5. Conclusionesp. 26
- Referenciasp. 27
- Tabla 1. Resumen análisis descriptivo variablesp. 18
- Tabla 2. Modelo Logitp. 19
- Gráfico 1. Curva probabilidad aciertap. 22
- Gráfico 2. Curva probabilidad plazop. 23
- Gráfico 3. Curva probabilidad ventasp. 24
- Gráfico 4. Curva probabilidad activop. 24
- Gráfico 5. Curva probabilidad cupop. 25