Methodological Foundations of The Paradigm of Digital Sustainability and Risk of The Banking System Based on The Use of Options (рart 2)

The article considers the theoretical foundations of the sustainability of the domestic banking system in the context of digitalization and increasing market uncertainty. The relevance of the study is due to the fact that in modern conditions there is a further increase in the scale of use of high-performance artificial intelligence systems. The scientific novelty lies in the fact that the study puts forward and proves the hypothesis that with the help of the artificial intelligence system DL-model «Random Forest» an accurate forecast of the closing price of Sberbank shares can be obtained, and also explores approaches aimed at filling the gap in theoretical issues related to modeling the mechanism for ensuring the sustainability of the banking system using options. The practical significance is that the results can be recommended for use in practice. The work used such research methods as statistical, monographic, analytical, the deep learning model «Random Forest» based on its method «DecisionTreeRegressor» (Decision Tree Regression), methods for constructing histograms in the Collab service, as well as Python libraries: pandas, sklearn and others. The neural network model included parameters reflecting the development of the banking system for the period from January 2016 to July 2024. The model showed good forecasting accuracy, so the forecast for November 1, 2024 was 272.07 rubles, with the actual value of the price of an ordinary share of Sberbank of 255.98 rubles. The forecast error was 16.03 rubles, or 5.91 % of the actual price. At the same time, the average absolute error was 11.64.