Modele prognozowania bankructwa w systemie wczesnego ostrzegania przedsiębiorstw
Bankruptcy prediction models in the corporate early warning system

Summary/Abstract: Warning predictions and early warning systems are, by their very nature, dynamic. Bankruptcy prediction models, on the other hand, are usually classification models and as such are static. For this reason, they do not constitute cohesive components of early warning systems. The paper is a review of major bankruptcy prediction models and it presents the measures of prediction efficiency based on classification models as well as the possible sources of errors in predicting corporate bankruptcy. The author stresses the necessity of considering business cycles as the major determinants of bankruptcy risk. The final part of the paper discusses the possibility of a dynamic approach to bankruptcy prediction models in artificial neural network, logit and SEM models.

• Details
• Contents