A microeconomic model for technical analysis signals Giulia Rotundo†, Marcel Ausloos ‡ †Department of Business, Technological, and Quantitative Studies, University of Tuscia, Viterbo, Italy. e-mail: giulia.rotundo@uniroma1.it ‡SUPRATECS & GRASP, Department of Physics, University of Liege, B-4000 Liege, Belgium. e-mail:marcel.ausloos@ulg.ac.be March 15, 2005 Abstract Mathematical and statistical tools for quantitative analysis of financial markets have been experienced a fast and wide growth that overcome technical analysis. However the latter has been used for a long time, still now books are published about it and there exist web pages that use it for providing trading suggestions. Moreover some intuitions like the Elliot waves have been shown to be encapsulated in a more complete theoretical framework. The aim of this paper is estimate the probability of the occurrence of signals used in technical analysis in a proper time scale from the aggregate behavior of a microeconomic model of heterogenous agents. 1 Introduction Quantitative analysis of financial market data has well assessed several prop- erties like the long term memory in volatility [4, 5, 7, 18], returns [8, 9, 10], speculative bubbles [21], and the presence of fractals [2] that has been exten- sively studied since the pioneering paper [19]. Many mathematical and statistical models [11, 12, 17, 23, 24] are available nowadays for a phenomenological description of financial data [1]. However theories derived from complex system can explain the aggregate behavior of markets through the analysis of its components at the microeconomic level, thus providing an important insight about the mechanism of the formation of financial quantities. The main characteristic that compose several microeconomic models of fi- nancial market is the presence of two main type of agents: the fundamentalists and the chartists [6, 13, 14, 15]. Some other models take into account also the presence of noise traders that act without market information thus creating 1