FORECASTING THE DEVELOPMENT OF BEHAVIOURAL FINANCE

Abstract

Introduction. In the process of aligning the interests of market participants, the behavior of economic agents is highly uncertain and subject to significant fluctuations, with varying degrees of irrationality and dependency on different economic cycles. Purpose. The author emphasizes the importance of understanding and forecasting future actions at the state, enterprise, investor, and individual levels to minimize economic risks, enhance capital reproduction, optimize resource utilization, and ensure the long-term sustainability of the economy and economic entities. It is necessary to assess the change in the emotional component of decision-making at different stages of economic development and identify the factors influencing it at the individual, firm, and investor levels. Furthermore, we need to forecast future changes and develop an evaluation algorithm for behavioral finance based on a combination of neuropsychology, behavioral finance, and artificial intelligence. Methods. The study included comparative methods to identify common characteristics of irrational decision-making in both stock markets and various stages of economic development. Analytical methods were used to determine changes in the functioning of the irrational sector during different economic phases, as well as to analyze the behavior of groups and individuals. Correlation and regression analyses were conducted to establish relationships between variables. Analogies were used to simplify understanding of the peculiarities of irrational behavior within groups and their individual members. Finally, a model was developed using artificial intelligence to collect, process, and forecast behavioral information for individuals. Results. The primary objective of this paper is to forecast irrational human behavior using artificial intelligence. This study develops a forecast of the impact of AI on human decision-making and introduces a novel method for evaluating and predicting human decisions leveraging AI and AI Recognized Behavior (AIRB) technologies. The paper delves into the characteristics of AIRB and outlines its operational algorithm. Factors influencing cognitive state have been identified. The study has determined the change in the irrational component at various stages of a country's economic development and stock market cycles. The relationship between types of moods and the level of irrationality has been established. The study has revealed a paradox and specific features of decision-making at the group level compared to individuals. The effectiveness and problems of forecasting group behavior have been assessed. The foundations for minimizing irrational behavior at the level of groups, enterprises, and individuals have been identified. Conclusion. The level of irrational behavior among economic agents varies depending on different stages of the economic cycle. The analysis of irrationality is detailedly modeled in the trading of stock market traders, where each candlestick, grouped into patterns, symbolizes various moods, confidence, distrust, calmness, greed, or emotional chaos. Reflecting investors' behavior on the behavior of economic agents, it is relevant to forecast the level of irrationality based on emotional behavior at different stages of the economic cycle. Thus, during the transition from economic recession to economic growth, one can observe such stages as denial, anger, bargaining, depression, and acceptance. Each stage characterizes a certain level of irrational behavior from high to minimal, and therefore affects the quality of forecasts. Economic stability also negatively affects the level of rationality; for example, constant economic growth leads to euphoria, reducing the quality of logical decisions. Each level of irrationality directly influences its boundaries. Thus, usually at a low economic stage and, accordingly, a high level of irrationality, there is a large share of the shadow economy, as confirmed by the analysis of the shadow economy level in countries by development level and the level of gambling among the population. Forecasting the development of artificial intelligence, we propose our own AIRB method for forecasting the level of human behavior, which is based on collecting information, processing historical decisions, and comparing them with decisions of other participants during different cognitive states. At the enterprise level, the quality of the group serves as a minimization, therefore, it is necessary to introduce a system for evaluating an individual based on the level of characteristics to determine compliance with the group. Additionally, minimizing the risk of data change by storing it in blockchain systems.