Bankruptcy forecasting: An empirical comparison of AdaBoost and neural networks☆
Introduction
Predicting corporate failure is a hot topic in management science due to its importance for making correct business decisions. The accuracy of the forecasting model is clearly of crucial importance in failure prediction because many economic agents not only enterprises but financial institutions, auditors, consultants, policy makers or clients are affected by the bankrupt of a firm. In classification terms, the type I error is especially important, i.e. when a firm which will fail in the future is classified as healthy. Owing to this fact many researchers have focused their effort on finding the most efficient classifier. In the last decades artificial neural networks have received special attention and several studies have dealt with failure forecasting using this technique. Here we present some of them only as examples. Wilson and Sharda [55] used a sample of 129 firms, 65 of which went bankrupt between 1975 and 1982 and 64 non bankrupt firms matched on industry and year. They applied resampling techniques to generate the training and test data sets and reached very satisfactory results with only five accounting ratios.
Serrano–Cinca [48] provided a data set made up of 66 Spanish banks, 29 of them in bankruptcy and the rest solvent. He used nine ratios chosen from amongst those most commonly employed in accounting empirical research. The author proved the superiority of the neural network model against linear discriminant analysis using the leaving one out estimation of the error. Charalambous et al [16] applied several neural networks methods to a dataset of 139 matched-pairs of bankrupt and non-bankrupt U.S. firms for the period 1983–1994. The authors compared the predictive performance of five methods, namely Learning Vector Quantization, Radial Basis Function, Feedforward networks that use the conjugate gradient optimization algorithm, the back-propagation algorithm and the logistic regression.
In this research, the neural network approach is compared to AdaBoosted [19] classification trees for predicting corporate failure. As far as we are aware, this is the first study to compare AdaBoost and Neural Networks capabilities for corporate failure prediction. To illustrate its usefulness, we will apply AdaBoost on a selection of Spanish companies, and in order to ensure that these results are general and can be projected to other European countries and to the United States, we will use financial ratios that have proved significant for predicting business failure in previous studies (e.g. Frydman [23]).
The lack of a unified theory on corporate failure has meant that most studies dealing with distress prediction have focused on increasing the accuracy of the model and have not always paid enough attention to the model interpretation. This is clearly important in failure prediction as the firm must make appropriate decisions. Ensemble methods like AdaBoost do not improve model interpretation by themselves. Even more, they break the model interpretation conveyed by a decision tree. But, on the other hand, attribute importance methods can be devised to provide useful information for problem understanding. We will also calculate a novel measure for the importance of variables to facilitate model interpretation. This measure takes into account how often variables are actually used in the individual trees and, on the basis of this measure, the variables can be ranked in terms of importance.
The following factors should be taken into account within the empirical application. We use the legal definition of corporate failure which only includes bankrupt and temporary receivership firms. This is the most common definition in corporate failure prediction literature. One numerical (the firm size) and two categorical variables (activity sector and legal structure) are included as descriptors in addition to the usual financial ratios. The AdaBoost method is applied to the failure prediction, analyzing the extent to which this methodology is suitable for the subject.
In Section 2 of this paper, we present the AdaBoost method included in the study with a discussion of how it works in practice and we describe the algorithm used. The following sections introduce the failure prediction problem and the data used in the analysis. The classification results are then presented and the well-known neural network model is compared with the novel AdaBoost classifier. Finally, following on from the empirical analysis, we present our conclusions.
Section snippets
AdaBoost
A classifier system builds a model which is able to predict the class of a new observation given a data set. The accuracy of the classifier will depend on the quality of the method used and the difficulty of the specific application [24]. If the obtained classifier achieves a better accuracy than the default rule, then the classification method has found some structure in the data enabling it to do so. AdaBoost [19] is a method that makes maximum use of a classifier by improving its accuracy.
Problem description
Predicting corporate failure is an important management science problem and its main goal is to differentiate those firms with a high probability of distress in the future from healthy firms. In other words, a model is built to forecast the moment of distress so that the firm's economic agents may make suitable decisions. In order to be able to predict failure, it is essential to have access to information about the company's situation. This information is basically given by financial ratios
Data description
The companies used in this study were selected from the SABI database of Bureau Van Dijk (BVD), one of Europe's leading publishers of electronic business information databases and provider of the Wharton Research Data Services. SABI covers all the companies whose accounts are placed on the Spanish Mercantile Registry. In the case of failed firms, firms which had failed (bankruptcy and temporary receivership) during the period 2000–2003 were selected, but with the additional requirement that
Experimental results
In this paper, the same failure prediction problem is solved using two different classification methods in order to compare their classification accuracies in this task. To estimate the real accuracy, the total initial sample of 1180 Spanish companies was divided into two sets: eighty percent were used as a training set to build the classifier, and the rest were hidden from the classification method and were presented as new data to check the prediction accuracy. The training set therefore
Conclusions
In this study, two classification methods have been compared, showing the improvement in accuracy that AdaBoost achieves against the Neural Network. As has been seen, AdaBoost is based on building consecutive classifiers on modified versions of the training set which are generated according to the error rate of the previous classifier, while focusing on the hardest examples of the training set. In the practical application, the legal concept of corporate failure have been used which includes
Esteban Alfaro Cortés teaches Statistics at the Faculty of Economic and Business Sciences in the University of Castilla-La Mancha. He completed his degree in Business in 1999 and got his Ph. D. in Economics in 2005, both in the University of Castilla-La Mancha. His thesis dealt with the application of ensemble classifiers to corporate failure prediction. Current research deals with spatial statistics and the combination of classifiers (decision trees and neural nets) for solving heated topics
References (58)
- et al.
Adabag: implements AdaBoost.M1 and bagging. R package version 1.0
Financial ratios, discriminant analysis and the prediction of corporate bankruptcy
Journal of Finance
(1968)- et al.
Corporate distress diagnosis: comparison using linear discriminant analysis and neural networks (the Italian experience)
Journal of Banking and Finance
(1994) Bankruptcy prediction for credit risk using neural networks: a survey and new results
IEEE Transaction on Neural Networks
(2001)- et al.
A comparison of ensemble creation techniques
- et al.
Bankruptcy prediction for credit risk using an auto associative neural network in Korean firms
- et al.
Predicting the out come following bankruptcy filing: a three state classification using NN
International Journal of Intelligence Systems in Accounting, Finance and Management
(1997) - et al.
An empirical comparison of voting classification algorithm: bagging, boosting and variants
Machine Learning
(1999) Financial ratios as predictors of failure
Neural nets or the logit model? A comparison of each model's ability to predict commercial bank failures
In International Journal of Intelligence Systems in Accounting, Finance and Management
(1997)
Neural networks for pattern recognition
Bagging predictors
Machine Learning
Arcing classifiers
The Annals of Statistics
Classification and regression trees
Hybrid neural network models for bankruptcy predictions
Decision Support Systems
Comparative analysis of artificial neural network models: application in bankruptcy prediction
Annals of Operation Research
Ensemble methods in machine learning
Application forecasting with neural networks. An application using bankruptcy data
Information and Management
Experiments with a new boosting algorithm
A decision-theoretic generalization of on-line learning and an application to boosting
Journal of Computer and System Sciences
Boosting the margin: a new explanation for the effectiveness of voting methods
The Annals of Statistics
Additive logistic regression: a statistical view of boosting
The Annals of Statistics
Introducing recursive partitioning for financial classification: the case of financial distress
Journal of Finance
Discrimination and classification
Neural networks. A comprehensive foundation
Bankruptcy analysis with self-organizing maps in learning metrics
IEEE Transaction on Neural Networks
Predicting bankruptcies with self organizing maps
Neurocomputing
Combining pattern classifiers. Methods and algorithms
A neural network for classifying the financial health of a firm
European Journal of Operational Research
Cited by (234)
A review of ensemble learning and data augmentation models for class imbalanced problems: Combination, implementation and evaluation
2024, Expert Systems with ApplicationsConstruction and analysis of the financing risk network of Chinese fisheries enterprises
2024, Ocean and Coastal ManagementDemand forecasting for fashion products: A systematic review
2024, International Journal of ForecastingInterpretable cost-sensitive regression through one-step boosting
2023, Decision Support SystemsThe evaluation of bankruptcy prediction models based on socio-economic costs
2023, Expert Systems with ApplicationsDevelopment and application of a hybrid forecasting framework based on improved extreme learning machine for enterprise financing risk
2023, Expert Systems with Applications
Esteban Alfaro Cortés teaches Statistics at the Faculty of Economic and Business Sciences in the University of Castilla-La Mancha. He completed his degree in Business in 1999 and got his Ph. D. in Economics in 2005, both in the University of Castilla-La Mancha. His thesis dealt with the application of ensemble classifiers to corporate failure prediction. Current research deals with spatial statistics and the combination of classifiers (decision trees and neural nets) for solving heated topics in the Economics.
Noelia García Rubio teaches Statistics at the Faculty of Economic and Business Sciences in the University of Castilla-La Mancha. She got her degree in Economics at the University of Madrid (UAM) in 1996 and completed her Ph. D. in Economics in 2004 on the construction of an intelligent and automated system for property valuation through the combination of neural nets and a geographic information system (GIS). Current research deals with spatial statistics and the combination of classifiers (decision trees and neural nets) for solving heated topics in the Economics.
Matías Gámez Martínez teaches Statistics at the Faculty of Economic and Business Sciences in the University of Castilla-La Mancha. He got his degree in Mathematics at the University of Granada in 1991 and finished a Master in Applied Statistics a year after. He completed his Ph. D. in Economics at the University of Castilla-La Mancha in 1998 on the application of geo-statistical techniques to the estimation of housing prices. Current research deals with spatial statistics and the combination of classifiers (decision trees and neural nets) for solving heated topics in the Economics.
D. Elizondo received the B.Sc. degree in computer science from Knox College, Galesbourg, IL, in 1986, the M.Sc. degree in artificial intelligence from the University of Georgia, Athens, in 1992, and the Ph.D. degree in computer science from the Universite Louis Pasteur, Strasbourg, France, and the Institut Dalle Molle d'Intelligence Artificielle Perceptive (IDIAP), Martigny, Switzerland, in 1996. He is currently a Senior Lecturer at the Centre for Computational Intelligence of the School of Computing at De Montfort University, Leicester, U.K. His research interests include applied neural network research, computational geometry approaches towards neural networks, and knowledge extraction from neural networks.
- ☆
Work partially supported by the Spanish Government under grant TIN2006-07262 and by the Castilla-La Mancha University under grants TC20070075 and TC20070095.