Analysis of Stability of The Banking System with The Hurwitz Criterion and Forecasting by The Random Forest Deep Learning Model

The article conducts a study of theoretical approaches to assessing the stability of the domestic banking system and forming a model that allows predicting its development in modern conditions. The relevance of the study is that in conditions of market uncertainty, approaches are increasingly being used to ensure the sustainable development of the banking system using artificial intelligence systems. During the study, pairwise correlation coefficients were identified between 11 factorial characteristics and the resulting characteristic — profi t growth (%), the Hurwitz matrix was formed. The determinant of the stability matrix of the banking system was calculated using the Hurwitz criterion. A DL-model "RF Regression" (DL-model "Random Forest") has also been formed, which allows predicting the increase in profi t of the banking system. The DL model was generated on the Collab service using the pandas, GridSearchCV, sklearn libraries using the results of the work of 144 commercial banks at the end of 2023. To calculate the determinant of a square matrix of dimension 11x11, a special program was written using the Python language. The calculated parameters of the DL model made it possible, based on the values of the regression equation coeffi cients, to form a polynomial and assess the stability of the banking system using the Hurwitz criterion. The resulting calculated value of the determinant of the Hurwitz matrix turned out to be greater than zero, which means that the banking system is stable in 2023. The novelty lies in the fact that the authors proposed approaches that make it possible to identify existing patterns in the development of the banking system, obtain a forecast of the dynamics of the banking system’s profi t and assess its stability using the Hurwitz criterion. The paper puts forward and proves the hypothesis that using the Random Forest DL model, a forecast of the increase in net profi t of commercial banks can be obtained, which predetermines the stability and dynamics of their development. The conclusions are that the results obtained can be successfully used in practice to predict and ensure sustainable development of the domestic banking system.