Since the first bankruptcy prediction models developed in the 60th of 20th century numerous different models have been constructed through the world. These individual models for bankruptcy prediction have been created in different time and space using different methods and variables. During this period various statistical methods have been used starting with the most popular univariate, linear and multivariate discriminant analysis, logistic regression, probit regression, decision trees, neural networks, rough sets, linear programming, principal component analysis, data envelopment analysis, survival analysis and so on. Therefore, we aim to provide deep insight and analyse the bankruptcy prediction models developed in countries of Visegrad four, with the emphasis on methods applied and explanatory variables used in these models, and evaluate them through appropriate statistical methods. Specifically, cluster analysis to explore the differences between basic groups of financial indicators and designed clusters of explanatory variables. Based on the analysis of more than one hundred bankruptcy prediction models we can conclude the most used variables, which serves as a basis for further research and development of prediction models in Visegrad group countries. Three clusters were developed which representing various explanatory variables while these clusters differ from basic groups of financial indicators. According to detected clusters we recommend to choose the most frequently used variables from each created cluster. From the cluster one revenues from sales/total assets ratio; from the cluster two the construction of models should contain current ratio, and from the cluster three we recommend to use ROE. Also if we take into consideration the total frequency together with the constructed clusters we advise to use more variables from clusters two and three. Results of the provided study may be used not only by researchers and enterprises but also by investors during the construction of bankruptcy prediction models in conditions of an individual country.
Keywords: bankruptcy, bankruptcy prediction, variables, countries of Visegrad four.