Intellectual System Development to Forecast of Probability of Bank Bankruptcy

Although the problem was discussed in literature before, the specific peculiar properties of Russian banking system remain significant and yet unsolved. The methodic used in prognosis of financial stability and bankruptcy risks are mostly globalized and cannot be applied for the research of reasons that lead to failure of certain banks. Therefore, the question of bankruptcy remains unanswered and its possible consequences grow more and more, asking for new research and new forecast methods. At this moment, the precious data considering the banking market as whole and certain banks in particular is collected and it is possible to try to predict with certain preciseness the risk of bankruptcy based on liquidity indexes and macroenomic markers that demonstrate the context in which the bank operates. In this research, the model was developed based on neural network that was designed and trained on the collected data. This model was implemented into an intellectual system that, by user's request can identify and show future and current risks that the bank experiences. To keep all of its information actual, the system refreshes it from public sources.