The process of credit risk management in financial institutions has been revised in recent years. In this context, large banks have developed and implemented several new techniques for measuring borrowers credit risk. This research aims to develop a risk classification model to assess the credit risk of companies in the Brazilian market. The model was built based on a sample of publicly traded companies classified as solvent or insolvent during the period from 1994 to 2004. Logistic regression was used to develop the model. The independent variables of the model are financial ratios, calculated from the financial statements and used as proxies of companies economic and financial situation. The validation of the model was done using the Jackknife method and a ROC Curve. The results of the study indicate that the risk classification model developed predicts default events one year prior to failure with good level of accuracy. The results also indicate that financial statements contain information that allow for the classification of companies as probably solvent or probably insolvent.
Credit risk model; Default event; Publicly traded companies; Logistic regression; Financial ratios
