Search

This paper studies the relationship between corporate failure forecasting and earnings management variables. Using a new threshold model approach that separates samples into different regimes according to a threshold variable, the authors examine regimes to evaluate the prediction capacities of earnings management variables. By proposing this threshold model and applying it innovatively, this research reveals boundaries within which earnings management variables can yield superior corporate failure forecasting. The inclusion of earnings management variables in corporate failure models improves failure prediction capacities for firms that manipulate substantial earnings. Furthermore, an accruals-based variable improves predictions of failed firms, but the real activities-based variable improves predictions of non-failed firms. These findings highlight the importance of indicators of the magnitude of earnings management and the tools used to improve the performance of corporate failure models. The proposed model can determine the predictive power of particular explanatory variables to forecast corporate failure.

This paper analyzes the prediction performance of human resources (HR) variables in corporate failure modeling. We define corporate failure as a two‐phase process from financial distress to bankruptcy, so that we can determine the prediction power of HR variables along a firm's phase in the financial deterioration process. We demonstrate the use of HR variables and their application to a two‐phase corporate failure model, providing first evidence for the predictive power of HR variables. The experimental results, based on real‐world datasets from Belgium, show that HR variables used in conjugation with accounting‐based information improve the accuracy of prediction modeling. However, the predictive power of HR variables varies in different phases of corporate failure with better prediction accuracy during the initial symptoms of corporate failure (i.e., financial distress). Findings show that our proposed model predicted financial distress with 84.1%, whereas the accuracy decreased to 83.3% when predicting bankruptcy. Besides, they also show that, on average, the inclusion of HR variables improves the global accuracy of the prediction models of 3.8% and allows to decrease Type I error of 5%.

Purpose Corporate failure remains a critical financial concern, with implications for both firms and financial institutions; this paper aims to review the literature that proposes corporate failure prediction models for the twenty-first century. Design/methodology/approach This paper gathers information from 106 published articles that contain corporate failure prediction models. The focus of the analysis is on the elements needed to design corporate failure prediction models (definition of failure, sample approach, prediction methods, variables and evaluation metrics and performance). The in-depth review creates a synthesis of current trends, from the view of those elements. Findings Both consensus and divergences emerge regarding the design of corporate failure prediction models. On the one hand, authors agree about the use of bankruptcy as a definition of failure and that at least two evaluation metrics are needed to examine model performance for each class, individually and in general. On the other hand, they disagree about data collection procedures. Although several explanatory variables have been considered, all of them serve as complements for the primarily used financial information. Finally, the selection of prediction methods depends entirely on the research objective. These discrepancies suggest fundamental advances in discovery and establish valuable ideas for further research. Originality/value This paper reveals some caveats and provides extensive, comprehensible guidelines for corporate failure prediction, which researchers can leverage as they continue to investigate this critical financial subject. It also suggests fruitful directions to develop further experiments.

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.

Small and medium firms are highly dependent on banks to finance their business activities. Thus, banking relationship may be crucial to overcome financial difficulties and to ensure their continuity. Accordingly, this paper investigates the influence of banking relationship on SMEs failure. In particular, four measures that firms can control to build their banking relationships and, that resemble standard variables from the literature on bank/firms relationships are evaluated: the breadth of relationships (number of banks), the relationship length(relationship duration), the relationship proximity (bank-firm distance) and, the relationship form (type of bank). Applying a logistic regression to a unique sample of 4960 French SME firms over the period 2013-2016, we evidence that banking relationships have a significant role on the SMEs likelihood of failure. More precisely, we find that multibank relationships, working with a small bank and relationship length are significantly negative correlated with SMEs failure. The opposite effect appears in bank-firm distance, which increases the SMEs probability of failure. Additionally, a corporate failure prediction model was built based on both financial ratios and banking relationship variables. The performance of this model was compared to a model based solely on financial ratios as predictive indicators. The results indicate that banking relationship variables possess prediction power to failure and enhance the performance of corporate failure models. Consequently, our findings are important from a policy perspective to further comprehend the role that banks play on SMEs failure.

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.

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.

We examine the impact of two types of earnings management, accruals and real activities manipulation, on bankruptcy prediction models. Consider that the form of engaging on earnings management might characterize firms’ financial condition, we hypothesize and find that the special information extracted from earnings management models, which encompass potential manipulations, serves to improve the accuracy of bankruptcy prediction models. We further provide evidences that when potential accruals manipulation is measured, it allows an improvement of bankrupt firms, while real activities measures, especially sales manipulation, enhance the prediction improvement of non-bankrupt firms. Our finding suggests that failed firms are more likely to engage on accruals manipulation, while healthy firms do it on real activities manipulation. As a result, this study contributes to the literature by analyzing a novel explanatory variable, earnings management variable, to predict bankruptcy that can overcome financial alteration and can be applied to any type of firms.