ANN Model for Corporate Credit Risk Assessment

The paper proposes decision tools to be used by commercial banks in order to classify the companies applying for credits in good and bad creditors, based on the number of days of delay in payment. Probit regression and neural networks, applied to a sample of companies which delayed or not their credit reimbursements are used to orient the decision taken by the bank, consistent with its priorities, of either minimizing the risk, or enlarging the customer base.