DESIGN OF ASSESSMENT AND FORECASTING OF A COUNTRY’S FINANCIAL SECURITY IN A CHANGE MANAGEMENT CONDITIONS

Abstract

The competitiveness of any country depends directly on the efficiency of its financial sector. Moreover, the level of financial security determines a country’s ability to further develop on an innovative basis. An important need is to model the assessment and forecasting of a country’s financial security in order to make timely adjustments to government decisions and allocate resources to improve the country’s financial capacity. Methods of monitoring financial security, especially at the macro level, are underde-veloped in the academic world. The tools for selecting the indicators, which are needed to assess financial security, need to be deepened. In addition, in times of innovative discoveries and technologies, the methods of such assessment need to be progressive and unconventional. The materials for this article were statistical indicators that characterise the financial development of Ukraine. It is possible to assess the level of security in a timely manner and to calculate forecasts, using these indicators. The authors have proposed such a grouping of these indicators, which will fully cover the financial activity of the country. We propose to model the assessment and forecasting of the country’s financial security with the usage of artificial networks. Application package Matlab including the Network Data Manager program was used for making neural networks, their training and application for forecasting the dynamics of financial security indicators. This program allows us to optimally form and select neural networks, topology, activation and training algorithm. The assessment and forecast of financial security have shown the need for increased government attention to the allocation and accumulation of financial resources. According to our estimates, the banking security indicators are expected to stabilise in 2023–2025. The indicators of the ratio of bank loans to deposits in foreign currency, the share of foreign capital in the share capital of banks, the ratio of long-term loans to deposits, and the share of assets of the 5 largest banks in the total assets of the banking system will stabilize at the level of 2020. Return on assets will slightly decrease and become negative, while the liquid assets to short-term liabilities ratio will increase to 95.6%. The ratio of total public debt service and repayment to government budget revenues is projected to increase significantly. The proposed design for assessing and predicting countries’ financial security is based on the usage of artificial neural networks, which are particularly relevant in the present context. The modelling logic is quite userfriendly and understandable for those interested and can be used by government officials to systematically determine the level of a country’s financial security. The proposed design can also be adapted for use in other countries to determine the level of financial security and to adjust the existing financial development strategies of the state.