Kushnir O. Models of estimation of operational risks of a bank.

The dissertation is devoted to the study of theoretical and methodological provisions and practical recommendations on the assessment of operational risks, taking into account the distribution of economic capital for risk-incidents of the bank. A conceptual approach to the assessment of the bank's operational risks based on a set of economic and mathematical models based on the experience of foreign banking institutions and the provisions of the Basel Committee and adapted to the Ukrainian realities of the banking sector is proposed. This made it possible to determine the level of possible financial losses caused by these risks by banking institutions.The mathematical toolkit of distribution of economic capital for risk incidents of a banking institution with the purpose of increasing the efficiency of management of its operational risks has been improved, allowing to determine the bank's own funds sufficient to cover potential losses at a given level of operational risk tolerance for a particular time horizon. The models of operational risk assessment based on the AMA (Advanced Measurement Approach) approach have been improved and tested, allowing to specify and detail the calculations for certain lines of business of the bank (business lines) and for a certain type of operational losses (risk incidents) established for each of the directions.The toolkit for forecasting the magnitude of operational risks and economic capital of the bank based on the application of neural network modelling in conjunction with the construction of rational splines is proposed, which allows improving the analytical capabilities of the calculated forecasted values of operational risks and enabling them to form an optimal adaptive strategy based on them that does not require a priori information on the structure desired functional dependence.