Virtual decision maker for stock market trading as a network of cooperating autonomous intelligent agents

The idea of a machine that can learn from its own interactions with the world has been one of the driving forces behind artificial intelligence research since its inception (Turing, 1950). The motivation of this paper is to present and demonstrate the merits of a machine to assist a non-expert decision maker, in applying stock market hedging strategies that are typically used by experts. The machine, called a Virtual Decision Maker (VDM), provides the processing power to deal with the very high granularity of such strategies while offering higher flexibility in choosing the trading frequency. More specifically, the VDM is a network of cooperating intelligent agents' technologies that can exploit automated on-line trading services at any time and any place without the physical presence of the decision maker. At present, the VDM is developed in an Excel-VB environment with agents that cooperate to: (i) import the required stock market real time data; (ii) identify the opportunity of making a trade; (iii) formulate an appropriate strategy; and (iv) execute of the corresponding order on the fly. The design of the VDM takes as its main premise the technological advantage of reduced reaction time, as opposed to attempting to anticipate a given security's movement. Results indicate in a disturbing manner that, given expert-validated knowledge, decision-making by cooperating and negotiation intelligent agents could lead to higher returns than commonly used indexes. In the conclusion, the idea of full automation is discussed in relation to the decision maker's behavioural and the cognitive issues.