The prediction of the firm ' s future development and possible failure is one of the most important information needed for decision making by all stakeholders. Early detection of impending bankruptcy creates the possibility of adopting remedial measures that can, if they are effective, avert it. The crisis in recent years has raised the focus on predictive models and their reliability. In response to the experience with the elder models and their lower reliability the researches aimed to construct new versions of the older models as well as the new models based on the wider scale of variables. One of the stream of researches is aimed to construct specific models for the condition of emerging economies, i.e. Polish, Slovak, Lithuanian and Czech environment. The aim of this paper is to compare the reliability of selected foreign bankruptcy models in the conditions of the Czech economy compared to the Czech model and to reveal if there are any differences in the prediction ability depending on the conditions under which they were compiled. We used a set of 80 Czech companies operating on the Czech market, in which insolvency was declared in 2017. Based on these data we compare the prediction ability of the selected models which were created in the foreign economies, both traditional economy and emerging. The results did not confirm the greater reliability of the model that was created in national conditions. The most reliable model has been identified the one created in terms of another transition economy.
