Microstructure Dynamics and Agent-Based Financial Markets: Can Dinosaurs Return?

摘要: This paper formalizes observations made under agent-based artificial stock market models into a concrete hypothesis, which is called the Dinosaur Hypothesis. hypothesis states that behavior of financial markets constantly changes and trading strategies in need to continuously co-evolve with it order remain effective. After formalizing we suggest testing methodology run tests 10 international markets. Our are based on framework recently developed, used Genetic Programming as rule inference engine, Self-Organizing Maps clustering machine for above rules. However, an important assumption study was maps among different periods were directly comparable each other. In allow this happen, had keep same clusters throughout time our experiments. Nevertheless, could be considered strict or even unrealistic. paper, relax assumption. makes model more realistic. addition, allows us investigate depth dynamics test plausibility The results show indeed markets' changes. As consequence, market; if they do not, become obsolete dinosaurs.