Abstract
Credit risk management is a key issue for any company at anytime, but is especially important in the case of the banking industry. This fact is more than evident in times of financial crises, when financial institutions can suffer high losses due to unpaid credits. For this reason, international financial supervisors and authorities have forced banks to monitor their credit risk and this risk is a variable that is constantly under the scrutiny of all financial agents in the international markets.
There are currently several methodologies that aim to predict the default probability of debtors. Many of them use logit analysis to discriminate among debtors. New methodologies make use of neural networks or multicriteria methods. This paper presents a new proposal based on goal programming, which allows the judgement of experts to be incorporated into the model, as suggested by the Basel Committee. Our approach combines the objective information of financial variables with the subjective judgement of experts about the different relevance of these variables, so observing the Basel Committee guidelines.
