Journal of Applied Economic Research
ISSN 2712-7435
Concepts of Improvement of Indistinct and Neural Network Methods of Modelling of Bankruptcies at Risk Management by the Loan Portfolio of Bank
Biryukov A.N., Kasimova L.I.
Abstract
Abstract. Modelling of failures is highly relevant from the point of view of corporate management in the real and financial-banking sector. To achieve effective corporate governance appropriate information and analytical support should be created, i.e. a system of monitoring the risk of bankruptcy of organizations in terms of the use of intellectual information technology. The specified monitoring is an important tool of ensuring the economic security of activity of the organizations. Indeed, organizations must constantly monitor their financial situation and analyze their financial stability because uncertainty and instability are the crucial properties of the economic and political environment today. Uncertainty is market conditions which are simultaneously affected by an immeasurable number of factors of different nature and orientation. Instability of the external environment is manifested through the uncertainty of the directions of change and their high frequency. This article discusses the issues and presents the results of research on the management of credit portfolio of a bank with the use of neural network models that provide new opportunities for reducing risks at the stage of bankruptcy of organizations with different change dynamics of the financial-economic condition of borrowers. Improving the efficiency of management of the bank’s credit portfolio based on the neural network logistic iterative dynamic method and the Mamdani fuzzy method. Much attention is paid to comparison of dynamic models with many famous foreign and domestic quantitative static models and expert rating models and methodology regulated by the government of the Russian Federation. Proposals are made for the consolidated management of credit portfolio of the bank. The proposed neural network logistic iterative dynamic method is summarized in terms of increasing its predictive power under the conditions of insufficient information about the intermediate values of probability of the bankruptcy of the borrower. The proposed ideas are tested in computational experiments on the real data from construction companies of Russia. The findings of the study show that the neural network iterative dynamic method allows us to build extrapolation models of failures of organizations in managing the credit portfolio of the Bank.
<object id="ieooui" classid="clsid:38481807-CA0E-42D2-BF39-B33AF135CC4D"></object>
Keywords
Key words: bank; solvency of borrowers; neural network model; fuzzy model; loan portfolio; factor models of bankruptcy of the concept.
About Authors
DOI: http://dx.doi.org/10.15826/vestnik.2016.15.4.28
Download full text article:
~630 KB, *.pdf
(Uploaded
04.10.2016)
Created / Updated: 23 September 2021 / 23 September 2021
© Federal State Autonomous Educational Institution of Higher Education «Ural Federal University named after the first President of Russia B.N.Yeltsin»
Remarks?
select the text and press:
Ctrl + Enter
Portal design: Artsofte
Contact us
Rector's Office
Rector, Dr. Victor Koksharov
Tel. +7 (343) 375-45-03, e-mail: rector@urfu.ru
Vice-Rector for International Relations, Dr. Maxim Khomyakov
Tel. +7 (343) 375-46-27, e-mail: Maksim.Khomyakov@urfu.ru