A PREDICTION-BASED BEHAVIORAL MODEL FOR FINANCIAL MARKETS

approximation. A feedback model for financial markets is proposed, in which the control action is an agent’s decision based on his beliefs of the price dynamics and his behavior reflecting his attitude, such as risk aversion or risk preference. An adaptation mechanism is described and the condition for equilibrium is formulated as a, typically non-linear, fixed point problem for operators. A data driven stochastic approximation procedure is given for the on-line tuning of the predictor to achieve equilibrium. Simulation results are also presented.