Forecasting corporate bankruptcy with an ensemble of classifiers

Abstract

Prediction of corporate bankruptcy is a phenomenon of growing interest to investors, creditors, borrowing firms, and governments alike. Timely identification of firms' impending failure is really wanted. The aim of this research is to use supervised machine learning techniques in such an environment. A number of experiments have been conducted using representative machine learning algorithms, which were trained using a data set of 150 failed and solvent Greek firms. It was found that an ensemble of classifiers could enable users to predict bankruptcies with satisfying precision long before the final bankruptcy. © 2012 Springer-Verlag.