The purpose of the study was to identify factors that increase the likelihood of revocation of a credit institution’s license by the regulator and to build a predictive model. The authors used the basic techniques of descriptive statistics, that is, the primary processing of empirical data, their systematization and visual representation, the cluster analysis of data, the building of a logistic regression, probit model and random forest algorithm, as well as forecasting based on these models, using the method of committees of three available models.