Bankruptcy Prediction by Supervised Machine Learning Techniques

Abstract

It is very important for financial institutions which are capable of accurately predicting business failure. In literature, numbers of bankruptcy prediction models have been developed based on statistical and machine learning techniques. In particular, many machine learning techniques, such as neural networks, decision trees, etc. have shown better prediction performances than statistical ones. However, advanced machine learning techniques, such as classifier ensembles and stacked generalization have not been fully examined and compared in terms of their bankruptcy prediction performances. The aim of this chapter is to compare two different machine learning techniques, one statistical approach, two types of classifier ensembles, and three stacked generalization classifiers over three related datasets. The experimental results show that classifier ensembles by weighted voting perform the best in term of predication accuracy. On the other hand, for Type II errors on average stacked generalization and single classifiers perform better than classifier ensembles.

Modeling credit scoring using neural network ensembles

“…They are widely used in the literature (see e.g. Hsieh, 2005;Tsai et al, 2011;West et al, 2005). The information contained in these data sets is shown in Table I.…”

“…outliers) to reduce prediction errors during the training of a classifier (Hsieh, 2005). Related work has shown that advanced machine learning techniques outperform single classifiers (West et al, 2005;Tsai et al, 2011). For instance, Hsieh ( 2005) constructed a hybrid classifier by combining the k-means clustering method and a neural network model, which shows that the hybrid classifier can provide greater prediction accuracy than the single neural network model.…”

Ensemble Machine Learning Approach to Bankruptcy Prediction

“…It uses a huge number of decision trees as base classi ers to construct a forest as multiple decision trees or forests can perform better than one. Therefore, using combined classi ers or combined learned models, desirable results can be obtained [55]. The model will be trained using samples from the training set selected randomly and with random replacements in each attempt to make sure that each sample has been trained at random and the best-trained model will be chosen to contribute to the nal decision of model construction.…”