FUTURE RESEARCH DIRECTIONS

More work is needed to obtain trading behaviour that much more closely resembles the behaviour of real FX market traders. Despite the fact that the stylized facts of our agent-based models of trad­ers replicate some of the stylized facts of real FX market traders’ behaviour, it is essential to define more stylized facts regarding the FX market trad­ers’ behaviour in order to evaluate the accuracy of the trading agents’ behaviour in resembling the

collective behaviour of real traders.

Another obvious way in which our work can be extended is to conduct a microscopic analysis of the trading agents’ behaviour, in an attempt to understand, identify, and measure through simu­lations, the different forces driving the traders’ behaviour. This can be examining under different situations, with the intention of looking for condi­tions under which collapses occur in the market. Through simulations, one can study the leverage and order size effect on traders’ behaviour, and the pressure of margin calls on such behaviour.

Going even further, we could build dynamic maps of traders’ positions in the market. These would present a snapshot of the entry prices for all open positions held by the traders in a defined period of time for a given currency pair. The traders’ positions maps can provide a new view to spots price changes, trends and the dynamics of traders’ behaviour. Besides, some of the differ­ent forces driving the FX market can be defined and measured from these maps. Therefore, these maps can be used as indicator of traders’ reac­tions to price movements, expectations of market dynamics, and can measure pressure on prices due to positions’ unrealized profits and losses.

An attractive line of research is to study the interaction between various trading strategies, such as strategies based on directional-change events, with the aim of evaluating different adopted strategies by assessing the traders’ performance in the market. This in turn could help us answer a number of questions, which have arisen such as: which adopted trading strategy leads to success or failure under different circumstances? Which trading strategy is most efficient? Going even further, more works could be done in observing the results of the interactions between different traders. Such observations can be used as feedback to change the models of interaction.

One interesting dimension of research is to study price formation on FX markets, which are organized as market-makers markets. Our model of FX market trading activity can be easily modi­fied to generate artificial prices resembling the FX market time series of prices, with the objective of finding conditions under which the statistical properties of price series in the artificial market resemble those of real FX markets.

An interesting and promising avenue of re­search is to study price competition among market makers in the FX market. Modelling the interac­tions between various market-makers in the FX market could provide behavioural explanations of some current phenomena in the FX market. There is some interesting research already be­ing done with regard to modelling competition between market-makers (Kyle, 1984; Kyle, 1985; Biais, 2000; Bondarenko, 2001; Nishide, 2006). Unfortunately, these works do not deal with high-frequency markets, which could provide a deep insight into the strategic trading behaviour of market-makers.

11.