Abstract:

This article deals with the application of the neuro-fuzzy approach for predicting bankruptcies in the Czech Republic. The prediction is based on financial indicators of manufacturing companies from 2012–2016. The obtained data are used to design ANFIS, with the advantage of the possibility to model complex, dynamic and non-linear problems, when compared to statistical methods. Based on the established results it can be stated that the designed model is capable of revealing an enterprise prone to bankruptcy with a given probability and thus giving a picture of its current development.