Diseño de un modelo de calificación de crédito para la minimización del riesgo de crédito con el uso de Visual Basic en la Cooperativa de Ahorro y Crédito San Valentín, cantón Quito, parroquia Chillogallo
The Cooperative of Savings and Credit San Valentine conducts its operations since 2011 in the southern sector of the city of Quito, this institution currently doesn't have an expert credit score system that serves as a tool for decision making regarding placement of credits. In this sense we ar...
| Autor Principal: | Analuisa Aguiar, Armando Andrés |
|---|---|
| Otros Autores: | Paredes Ñacata, Adriana Elizabeth |
| Formato: | bachelorThesis |
| Idioma: | spa |
| Publicado: |
2015
|
| Materias: | |
| Acceso en línea: |
http://dspace.ups.edu.ec/handle/123456789/9450 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: |
The Cooperative of Savings and Credit San Valentine conducts its operations since 2011 in the southern sector of the city of Quito, this institution currently doesn't have an expert credit score system that serves as a tool for decision making regarding placement of credits. In this sense we are going to try to design a credit scoring model for this institution that minimizes its risk in the microcredit portfolio.
The credit scoring model begins with the identification of qualitative variables (text type) and quantitative (numerical type) these variables are obtained from the historical database of the Cooperative, the data obtained correspond to the microcredit portfolio.
In the first analysis we have to perform a Univariate exploratory analysis where the initial variables decrease. After that we have to perform a Bivariate exploratory analysis we must determinate if the remaining variables explain the dependent variable, this dependent variable is created with two initial variables named "mora promedio" and "mora máxima", this dependent variable is called "INCUMPLIMIENTO" and this variable identifies, discriminates and determines a GOOD or BAD client.
After the Bivariate exploratory analysis we got the final variables that explain significantly the dependent variable "INCUMPLIMIENTO" and we are going to use those final variables in a logistic regression in order to get a general equation of credit score model.
This equation gives a result known as Zj this value is the sum of the final variables but this result is not a probability for that reason we have to adjust this result in order to get a probability between zero and one (0≤Pi≤1) known as Pj.
Pj means a probability. If the value is closer to zero, the client will have a lower probability of default. If the value is closer to one, the client will have a higher probability of default. |
|---|