Simulation Modeling of Information Transfer Between Stock Exchanges Based on a Multi-Agent Model

Abstract

The article substantiates the relevance of solving the problem of stock exchange modeling as an information system accumulating the orders of market participants and executing them. The object of the study is the process of information transfer between stock exchanges. The subject of the study is the use of simulation multi-agent model to study the process of information transfer between stock exchanges. The paper analyzes the possibilities of building a simulation model to analyze rare phenomena in the financial market using a multiagent approach. It is substantiated that this approach allows taking into account those features of complex dynamic systems, for which either no effective analytical approaches have been developed or it is impossible to conduct the necessary numerical experiments. The proposed logic of the market simulator is based on the allocation of certain trader classes, that may differ in the logic of making their decisions on the stock exchange, for this purpose the scheme of agent interaction in the model of the stock exchange is described. On the example of modeling the information transmission and market shocks between markets the possibilities of the proposed simulator are studied. The use of the proposed model in practice allows us to analyze rare phenomena occurring on the stock exchange, which cannot be described analytically, but their recreation by means of the simulation model will allow us to study them. The study of rare phenomena on the stock exchange with the help of simulation modeling allows to obtain different variants of the results of the phenomena at different sets of parameters of the model (agents), i.e. to carry out parameter.