Consensus-based trading strategy in a-trader system

The authors of this paper present an approach to trading strategy design for an A-Trader multi-agent system which supports investment decisions on the stock market. The functionalities of the system, and the main component, the Supervisor Agent, used as a strategy a consensus method to reduce the level of investment risk, are described. The consensus method allows the coordination of the work of agents, and on the basis of the decisions provided by the agents presents trading advice to the investor. The strategy has been tested on FOREX quotes, namely on the pair USD/PLN. The results of the research are described and the directions of the further development of the platform are provided in the conclusion.