Price fluctuations in market model with heterogeneous trading strategies

We present a model of market's dynamics based on heterogeneous trading activities. The price evolution is calculated by simulation on a cellular automaton. Agents in the market trade according to fundamentalist and chartist strategies; and according to a self-learning mechanism dynamically connecting strategies with share holdings. As traders switch strategies they determine a price dynamics characterized by both stochastic drift and stochastic Wiener volatility. We analyze three market scenarios dominated by (1) fundamentalists (2) chartists and (3) neither. The model accounts for mean-reversion, bubble formation and burst phenomena.