Bank Default Probability Modelling

The purpose of the research is to create and compare the econometric model and the model based on neural networks to assess the probability of banks' default . The object of the study is the banking sector of the Russian Federation. The subject of the study are methods for assessing the probability of default of banks. The first chapter of the work is devoted to reviewing existing methods of assessing the probability of banks' default, comparing them and highlighting advantages and disadvantages. In the second chapter, a description of the initial data array and an analysis of used variables are presented. The third chapter describes the process of constructing econometric and neural network models for estimating the probability of banks' default, their comparison and Investigation of the dependence of the probability of default on the values of financial indicators using the constructed models. The research work includes 3 chapters. Scope of work: 66 pages.