The authors of this paper present the architecture of a multi-agent system which supports investment decisions on the stock market. The individual components of the system, the manner of communication between them, the mechanism of assessing the individual agents are discussed. New methods of pre-processing financial series, the concept of never-ending learning, the behavioural model of stock exchange traders and the manners of translating the modelled patterns into the open and close position signals are described. The results of the research are described and the directions of the further development of the platform are provided in the conclusion.
JavaScript is turned off in your web browser. Turn it on to take full advantage of this site, then refresh the page.