Agent-based simulation of herding in financial markets

There are several models of financial markets which look at the herding effect. This is a situation where many market traders act as a herd in that they all behave in a similar way with their trading. This type of behaviour may explain certain observed characteristics (or ‘stylised facts’) in real markets. However, the various models have different herding mechanisms and market settings This paper sets out the rationale of our approach and our initial work in trying to get a better understanding of herding in financial markets. Our research, though, is at an early stage. The basic methodology is to reproduce and compare some of the existing models, hopefully leading to a more general understanding and measure of herding and the relationship with market behaviour. One model has been investigated so far and this is described. A more general issue is the research importance of reproducing previous studies