Microscopic model of financial markets based on belief propagation

We present a simple microscopic model of financial markets based on belief propagation in order to simulate the dynamics of the stock markets. A two-dimensional small-world communication structure is introduced in our model and the beliefs of market leaders spread on the network which results in the herd behaviors of traders. Most of the stylized aspects of the financial market time series, including multifractal property, are reproduced by the model. A direct comparison is made with the daily closures of the Shenzhen composite index.