The review of logit-regression models predicting bankruptcy

The article deals with the current issues of bankruptcy predicting; shows the classification of bankruptcy prediction models according to the criterion of the applied approach. On the basis of the study of domestic and foreign authors ' scientific researches, it presents comparative characteristics of models received when using a multiplicative discriminant analysis and logistic regression. The paper describes the characteristics of existing logit-regression models, developed in various countries, it establishes the effectiveness of the logistic regression unit and defines an algorithm for constructing a logit-regression model. In conclusion, the work notes that using of the same mathematical apparatus on a sample of Russian companies and their indices, calculated according to accepted accounting standards, it is possible to obtain an accurate model for predicting the bankruptcy of domestic enterprises.