An examination of misclassifications with bank failure prediction models

Unlike prior work in the area of bank failure prediction, this article focuses on misclassifications: the individual banks that were predicted by a model to fail and yet have not, and those predicted to survive and yet have failed. The concern here is with the profile of these misclassified banks and the causes of their success or failure. Linear and quadratic multiple discriminant analysis models and Cox proportional hazards models are employed. The performance of these models is examined in periods subsequent to that over which they were estimated. The results provide insights for the development and application of early warning models for banks.