The purpose of this study is to present the Cox proportional hazards model and to apply this model to the prediction of bank failures. The Cox model, which has been used extensively in biomedical applications, has not been previously employed in the finance literature. The principal advantage of the Cox model over other classification techniques is that it models the expected time to failure. Results of the study indicate that total classification accuracy of the Cox model is similar to that of discriminant analysis, although the Cox model produces somewhat lower type I errors. In a comparison of actual and predicted times to failure, the Cox model tends to identify bankruptcies prior to the actual failure date.