This chapter presents a market microstructure model, which investigates the behavior dynamics in financial markets. We are especially interested in examining whether the markets’ behavior is non-stationary, because this implies that strategies from the past cannot be applied to future time periods, unless they have co-evolved with the markets. In order to test this, we employ Genetic Programming, which acts as an inference engine for trading rules, and Self-Organizing Maps, which is used for clustering the above rules into types of trading strategies. The results on four empirical financial markets show that their behavior constantly changes; thus, agents’ trading strategies need to continuously adapt to the changes taking place in the market, in order to remain effective.
Natural Computing in Computational Finance Studies in Computational Intelligence Volume 380, 2012, pp 181-197