Investment analysis & decision making in markets using adaptive fuzzy causal relationships

Financial markets constitute an emerging area for design and application of computational intelligent decision support tools. From numerous research studies, the main conclusions drawn refer to the existence of boundaries on current forecasting capabilities, paired with the need for innovative modeling frameworks structured on benefit/risk principles. As such, the idea to approximate the investors' reasoning process seems to be a promising direction for decision analysis and forecasting purposes. To underpin such a scheme, Fuzzy Cognitive Maps (FCM) constitute an interesting mathematical/simulation alternative. Thereby, this methodology is exploited to build a stock price forecasting system. In this work, however, proportionate significance sustains the theoretical and practical modifications, in order to enhance the adaptive potential and simulation effectiveness of FCM-based models. Particular emphasis is given on the capability of FCMs to capture non-linear causal propagations and to simulate complex environmental phenomena, like the inertial and multistimulus forces exhibited by a plethora of physical and technical systems.