Examinando por Materia "Credit scoring"

Mostrando 1 - 8 de 8
  • No hay miniatura disponible
    Publicación
    Creación de una app Fintech de crédito por descuento de nómina para la empresa Créditos e Inversiones de Colombia (CICOLSAS), como estrategia de crecimiento empresarial
    (Universidad EAFIT, 2022) Valencia Tovar, Diana Marcela; Gómez Guerrero, David Wady; Durango Gutiérrez, María Patricia
    This research was developed with the aim of designing a fintech app that offers credit through payroll deduction, for Créditos e Inversiones de Colombia (Cicolsas), Colombian Credits and Investments (in English). To do so, descriptive-analytical research was conducted in three steps: (1) definition of the app structure and content, (2) design of a tool to measure credit risk for the app (using a logistic regression model), and (3) identification of the necessary capital structure and its cost (through a CPPC methodology). As a result of the research, it was possible to develop an app with a landing page and nine information modules. Three types of users can get access to it: registered clients, non-registered clients, and administrators. It was integrated a credit scoring into the app, built with the historical information provided by Cicolsas. This credit scoring has sixteen predictive variables that allow the assessment of new clients’ default likelihood. Finally, an investment proposal of 1000 million COP has been established, funded with third-party and the company’s own resources, in a 50/50 proportion. This investment proposal has a global capital cost equivalent to 9,078% E.A.
  • No hay miniatura disponible
    Publicación
    Desarrollo y evaluación de un modelo de scoring para el otorgamiento de crédito a poblaciones vulnerables : caso fintech del instituto para el Desarrollo de Antioquia (IDEA)
    (Universidad EAFIT, 2024) Maya Osorio, Karla María; Osorio Ramírez, John Alexander; Álvarez Franco, Pilar; Cruz Castañeda, Vivian
  • No hay miniatura disponible
    Publicación
  • No hay miniatura disponible
    Publicación
    Implementación de un modelo de scoring de crédito para Mexichem Colombia SAS
    (Universidad EAFIT, 2024) Vargas Izquierdo, Daniel Camilo; Rojas Ormaza, Brayan Ricardo
    This project sought to implement a credit scoring model for Mexichem Colombia SAS using machine learning techniques to predict the probability of default in companies. Four algorithms were compared: decision trees, random forest, gradient boosting and neural networks, each with unique characteristics in terms of accuracy and handling of complex data. The research included the selection and evaluation of relevant variables using the Gini index and recursive elimination techniques to avoid overfitting. The results helped to identify the most effective model to predict credit risks, optimizing financial decision-making.
  • No hay miniatura disponible
    Publicación
    Modelo de credit scoring para la empresa Grupo Factoring de Occidente S. A. S.
    (Universidad EAFIT, 2021) Urán Vélez, Alejandro; Santiago Sandoval, Carolina; Rojas Ormaza, Brayan Ricardo
    Credit risk consists of the establishment of policies and procedures, wish are appropiate to current regulations and to the shareholder’s risk profile and tend to minimize the probability of occurrence of situations that put the companie´s resources at risk through an adequate percentage of portfolio provision due to default of payers. The present research aims to contribute to the development of a credit scoring model that allows to do tracking to the company´s customers “Grupo Factoring de Occidente SAS” to analyze the risk of payment failures, using the most relevant and significant variables of both the clients, the sector and the environment, through the use of previously proven econometric models.
  • No hay miniatura disponible
    Publicación
    Modelo de scoring para una entidad financiera especializada en el otorgamiento de crédito de vehículos
    (Universidad EAFIT, 2022) Astudillo Girón, Valeria; Rojas Ormaza, Brayan Ricardo
    This paper presents the design of a credit scoring model at the stage of granting credit to an entity specialized in vehicle financing, to identify the probability of default of creditors effective until April 2020, including the most relevant variables according to the methodology developed. This stems from the need of the entity to implement practices that allow it to control possible losses in its assets, taking into account that credit scoring are statistical models that support the management of credit risk in its measurement and monitoring
  • No hay miniatura disponible
    Publicación
  • Miniatura
    Ítem
    Supervised Statistical Methods to Identify Credit Acceptance Rate
    (Universidad Eafit, 2021-04-10) Yusty, Valentina; Laniado, Henry; Universidad Eafit, School of Sciences, Department of Mathematical Sciences
    Since incorrect decisions can have detrimental effects on financial institutions, the possibility for these to forecast business failures becomes indispensable. In the financial domain, the focus of research problems rarely revolves around the identification of the clients who desist their credit offering, but rather on bankruptcy prediction and credit scoring. The general objective of this paper revolves around the implementation of supervised machine learning algorithms that will allow CrediOrbe, a credit company, to target customers whose profile assimilates those who desist their credit offering. Machine learning algorithms have been greatly studied as tools to aid decisions makers in the realm of finance. Performance measurements are calculated and analyzed through the use of statistical classification measurements. Suggestions for further research are provided