The concepts, principles and procedures for developing and implementing a credit-scoring model had been fully developed by the early 1970s. However, the acceptance of these methods by loan officers was not great because they usually applied some relatively subjective evaluation procedures using a heuristic rather than scientific criteria. Nevertheless, the increasing lending institutions worry for diminishing the unpaid loans rate has led them to apply the newest statistical techniques. Moreover, this mind change has allowed them to have a greater control over credit policies. In this sense, the main target of this study is to supply an useful tool capable of classifying good and bad loans by the analysis of firm creditworthiness in an objective way. This work will be performed in cooperation with a very important spanish lender institution. The idea is to use tree-based methods in order to improve the results provided by traditional credit scoring procedures.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados