Company financial distress prediction is one of the most discussed issues of economists around the world in recent decades. From the first attempts in the 1960s to the present, one of the most widely used method to create these models is Multiple Discriminant Analysis. In the paper, we present the prediction model for Hungarian companies created using this method based on real data from the financial statements obtained from database Amadeus. Our database contains data of more than 250,000 companies and 26 financial indicators used as predictors. There is possibility to predict the financial difficulties of companies one year in advance using this model.