Multicriteria Framework for the Prediction of Corporate Failure in the UK

This study investigates the efficiency of two multicriteria decision aid methods, namely UTADIS and MHDIS in the development of business failure prediction models in the UK, as opposed to models developed with discriminant analysis and logistic regression. The dataset consists of 200 manufacturing UK firms out of which 100 failed during the period 2001-2003. The models are developed and validated using 10-fold cross validation. The results show that UTADIS and MHDIS achieve satisfactory classification accuracies, while both outperform logistic regression and discriminant analysis. Thus, the developed MCDA models could be of particular interest to creditors, investors, auditors and regulators in the UK