David Veganzones

Les travaux récents de la recherche académique ont permis d’affiner les outils capables de prévoir les faillites d’entreprises. Les progrès portent sur la qualité des données, le choix des variables, la prise en compte de facteurs exogènes et du choix de l’échantillon. Les banques pourraient ainsi améliorer les modèles utilisés pour évaluer les risques de défaillance.

Previous studies of bankruptcy prediction in imbalanced datasets analyze either the loss of prediction due to data imbalance issues or treatment methods for dealing with this issue. The current article presents a combined investigation of the degree of imbalance, loss of performance, and treatment methods. It determines which imbalanced class distributions jeopardize the performance of bankruptcy prediction methods and identifies the recovery capacities of treatment methods. The results show that an imbalanced distribution, in which the minority class represents 20%, significantly disturbs prediction performance. Furthermore, the support vector machine method is less sensitive than other prediction methods to imbalanced distributions, and sampling methods can recover a satisfactory portion of performance losses. Accordingly, this study provides a better understanding of the data imbalance issue in the field of corporate failure and serves as a methodological guide for designing bankruptcy prediction methods in imbalanced datasets.

Abstract

Financial information has been widely used to design bankruptcy prediction models. All research works that have studied such models assume that financial statements are reliable. However, reality is a bit different. Indeed, firms may tend to present their financial accounts depending on particular circumstances, especially when seeking to change the perception of the risk incurred by their partners, and thus distort or alter some of them. Consequently, one may wonder to what extent such “manipulations”, called earnings management, may influence any model that relies on accounting data. This is why we study how earnings management may affect financial variables and how it can indirectly distort predictions made by failure models. For this purpose, we used a measure that makes it possible to assess potential account manipulations, and not effective manipulations. Our results show that when these distortions are measured and used with other financial variables, models are more accurate than those that solely rely on pure financial data. They also show that the improvement of model accuracy is essentially due to a reduction of type-I error—the costliest error in economic terms.

Abstract

This paper presents two new clustering techniques based on Extreme Learning Machine (ELM). These clustering techniques can incorporate a priori knowledge (of an expert) to define the optimal structure for the clusters, i.e. the number of points in each cluster. Using ELM, the first proposed clustering problem formulation can be rewritten as a Traveling Salesman Problem and solved by a heuristic optimization method. The second proposed clustering problem formulation includes both a priori knowledge and a self-organization based on a predefined map (or string). The clustering methods are successfully tested on 5 toy examples and 2 real datasets.

Abstract

This paper proposes a methodology for identifying data samples that are likely to be mislabeled in a c-class classification problem (dataset). The methodology relies on an assumption that the generalization error of a model learned from the data decreases if a label of some mislabeled sample is changed to its correct class. A general classification model used in the paper is OP-ELM; it also provides a fast way to estimate the generalization error by PRESS Leave-One-Out. It is tested on two toy datasets, as well as on real life datasets for one of which expert knowledge about the identified potential mislabels has been sought.

Actuellement, les banques utilisent des techniques de scoring pour déterminer la santé des entreprises. Le principe est simple : il s’agit de qualifier une entreprise au travers de ses états financiers et de déterminer une limite de risque au travers d’un indicateur synthétique portant sur le secteur dans son ensemble. On compare le résultat de l’entreprise avec cette limite pour savoir si à court terme l’entreprise va faire ou non faillite.