Default Prediction in Companies of Real Sector

In a last year the bankruptcy forecasting has attracted increasing attention of researchers, especially important it has become after the crisis of 2008-2010. The possibility of the company to predict their future financial condition allows it to change or correct policy of the actions for creation of a financial steady position in the company. The article considers foreign and russian research on the topic of forecasting of probability of bankruptcy (Altman, 1968; Ohlson, 1980; Evstropov, 2008; Demeshev and Tikhonova, 2014) where considered various methods of predicting bankruptcy and examines the factors that influence its onset (approach). The aim of this research is to analyze the predictive power of logistic regression and the importance of certain financial targets in a model of bankruptcy of Russian companies. The sample contains financial indicators for two groups of companies, namely companies bankrupt and financially prosperous companies. For achievement of the purpose will be constructed logistic regression and ROC-curve to determine the quality of the model. The results of bankruptcy forecasting are useful not only for the companies, but also their counterparties and credit institutions, which decide to loan.