Ідентифікація матриці суміжності у моделі імпульсних процесів з різнотемповою дискретизацією в когнітивній карті застосування криптовалют

• Datum: 2022
• 1. Verfasser: Kantsedal, Heorhii
• Format: Artikel
• Sprache: Ukrainian
• Veröffentlicht: V.M. Glushkov Institute of Cybernetics of NAS of Ukraine 2022
• Schlagworte: когнітивна карта; імпульсний процес; криптовалюта; різнотемпова дискретизація; ідентифікація
• Online Zugang: https://jais.net.ua/index.php/files/article/view/6
• Назва журналу: Problems of Control and Informatics

The stagnation of cryptocurrencies in the financial markets is characterized by a collapsible dynamic, as it reflects the non-stationarity of processes and the inconsistency of the situation. In the process of cryptocurrency usage, various developments are aimed at changing the level of up to the replacement of cryptocurrencies. That is why, during operations with cryptocurrencies, the onset of risks is blamed [1]: The risk of losing users leads to a decrease in the price of Bitcoin; Risks associated with incorrect general simultaneous expectations of many users, created by manipulations of traders on financial exchanges; The risk of a sharp collapse of the cryptocurrency exchange rate as a result of common manipulations on the exchanges, which can include the so-called high-frequency trading, which consists in the presence of a specific group of users in the presence of a speed advantage in buying monetary assets before the majority of investors and selling them to slower users until the information about the purchase reaches to the slow investor. These actions, in combination with algorithmic trading, the mechanism of derivatives, and quarterly futures implemented on the stock exchanges, create a real threat of a significant change in the exchange rate from relatively minor disturbances; Risks associated with the lack of a guarantee for the preservation of capital invested in the purchase of cryptocurrency, which leads to a certain hysteria among users in the process of trading on stock exchanges. To describe the impact of these risks, the article examines the cognitive map (CM) of the use of cryptocurrency in the financial market, based on which a dynamic model of impulse processes of CM is described in the form of systems of difference equations (Roberts equation [1]) with different rate discretization. At the same time, the decomposition of the initial theoretical model of impulse processes of CM with one-rate discretization into subsystems with rapidly measured and slowly measured coordinates of CM vertices was performed. For this, subsystem models are presented with different discretization coordinates and are interconnected. Algorithms for identifying the coefficients of the adjacency matrix of impulse processes of CM for subsystems are developed based on the recurrent least squares method, respectively, in fast-changing and slowly-changing time scales. Based on digital modeling, experimental studies of the speed and accuracy of estimating the weighting coefficients of adjacency matrices in models of impulse processes of CM subsystems were performed.