A data mining approach to the prediction of corporate failure

This paper uses a data mining approach to the prediction of corporate failure. Initially, we use four single classi®ers Ð discriminant analysis, logistic regression, neural networks and C5.0 Ð each based on two feature selection methods for predicting corporate failure. Of the two feature selection methods Ð human judgement based on ®nancial theory and ANOVA statistical method Ð we found the ANOVA method performs better than the human judgement method in all classi®ers except discriminant analysis. Among the individual classi®ers, decision trees and neural networks were found to provide better results. Finally, a hybrid method that combines the best features of several classi®cation models is developed to increase the prediction performance. The empirical tests show that such a hybrid method produces higher prediction accuracy than individual classi®ers.