SOLUTIONS AND RECOMMENDATIONS

Modelling the market traders’ behaviour involves major sequence steps. The initial step is to acquire a good knowledge of the behaviour and mecha­nisms of the financial market. Mainly understand the dynamics and the forces that drive the traders’ behaviour in the market.

In this chapter, we have identified two prop­erties that a well-functioning model of financial market should possess: simplicity and heterogene­ity. We believe that the simplicity description of the parameters involved in the model is essential. This is to avoid complexity that could prevent researchers from examining the effect of each of the model’s parameters in replicating the real market traders’ behaviour. In fact, by using this approach, we found a set of conditions under which some stylized facts of traders’ behaviour emerge. On the other hand, the role of heterogeneity in modelling the real traders’ behaviour is impera­tive as it is desirable to model the markets in a more realistic manner. Heterogeneity of all sorts and forms must be incorporated in modelling the market traders: preferences, demands, aspirations, trading time windows, budget constraints, trading strategies, etc.

The validation of the agent-based models of traders verifies if the stylized facts of real traders’ behaviour are exhibited. Agent-based models of traders must be capable of resembling, to a certain extent, the collective behaviour of the real market traders. There are a number of approaches that can be used to validate the agent-based models of traders.

The several parameters that one is allowed to define in an agent-based market results in nu­merous degrees of freedom, which raises some difficulties when modelling the real market. The appropriate selection of the market parameters is a crucial element of the design but, at the same time, is a tricky task. Researchers should con­duct experiments with different values for each parameter, and then track the changes in market behaviour. This allows them to define exactly the parameters that will allow the model to replicate some real market features. As noted by LeBaron (2001a), understanding exactly where the parameter boundaries are between simple and complex behaviours is crucial to understanding the mechanisms that drive agent based markets” (p. 258).

10.