Citas bibligráficas
Rodas, J., (2024). Aplicación de árboles de inferencia condicional y regresión logística para el desarrollo de modelos de Score para Behaviour de créditos para clientes de persona jurídica [Trabajo de suficiencia profesional, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/27514
Rodas, J., Aplicación de árboles de inferencia condicional y regresión logística para el desarrollo de modelos de Score para Behaviour de créditos para clientes de persona jurídica [Trabajo de suficiencia profesional]. PE: Universidad Nacional de Ingeniería; 2024. http://hdl.handle.net/20.500.14076/27514
@misc{renati/960963,
title = "Aplicación de árboles de inferencia condicional y regresión logística para el desarrollo de modelos de Score para Behaviour de créditos para clientes de persona jurídica",
author = "Rodas Mendoza, Juan Junior",
publisher = "Universidad Nacional de Ingeniería",
year = "2024"
}
In the last half of 2019 and the first half of 2020, the risk premium of the client portfolio of legal entities of a financial institution increased its risk premium by an average of 50%. In that period, the portfolio was managed according to the experience of the product specialists. An essential tool for credit risk management is credit behavior in the financial system. One way to summarize this behavior is through the generation of risk profiles created based on a model. In this sense, at the end of 2020, I developed models of score for Behavior where conditional inference trees and logistic regression were applied using variables of the Financial System, Financial Statements and internal financial information of the clients. The objective of the model was to identify clients, legal entities, who may have a risk of non-payment of their debt based on risk groups established based on score ranges, in order to manage the credit portfolio. By using the models for Behaviour, it was possible to reduce the risk premium by 30%.
Este ítem está sujeto a una licencia Creative Commons Licencia Creative Commons