Market microstructure: can dinosaurs return? a self-organizing map approach under an evolutionary framework

摘要: This paper extends a previous market microstructure model, which investigated fraction dynamics of trading strategies. Our model consisted two parts: Genetic Programming, acted as an inference engine for rules, and Self-Organizing Maps (SOM), was used clustering the above rules into strategy types. However, purposes experiments our work, we needed to make assumption that SOM maps, thus types, remained same over time. Nevertheless, this could be considered strict, even unrealistic. In paper, relax assumption. offers significant extension because it makes more realistic. addition, allows us investigate behavior. We are interested in examining whether financial markets' behavior is non-stationary, implies strategies from past cannot applied future time periods, unless they have co-evolved with market. The results on empirical show its constantly changes; thus, agents' need continuously adapt changes taking place market, order remain effective.