Credit risk management analysis: An application of fuzzy theory to forecast the probability of default in a financial institution

Resumen

The aim of this research is to model the credit risk, estimating the impact of operational risk and customer features, using the fuzzy version of the LOGIT model. For this purpose, it proposes a Fuzzy Financial Risk Management Model, composed of a Credit score estimated with a Fuzzy LOGIT model and a Fuzzy Triangular Value-at-Risk adjustment. For this purpose, the probability of default of 3,746 commercial loans was predicted. The results show that the proposed methodology recognises the relationship between credit and operational risk better than traditional models. In conclusion, the proposed model provides an assessment of risk and measures it in terms of interest rate basis points. In addition, it provides the expected loss for three degrees of uncertainty. Therefore, the proposed methodology provides a suitable support for the design of credit policies in a financial institution.