Learning is neither sufficient nor necessary: An agent-based model of long memory in financial markets

摘要

Financial markets exhibit long memory phenomena; certain actions in the market have a persistent influence on market behaviour over time. It has been conjectured that this persistence is caused by social learning; traders imitate successful strategies and discard poorly performing ones. We test this conjecture with an existing adaptive agent-based model, and we note that the robustness of the model is directly related to the dynamics of learning. Models in which learning converges to a stationary steady state fail to produce realistic time series data. In contrast, models in which learning leads to continuous dynamic strategy switching behaviour in the steady state are able to reproduce the long memory phenomena over time. We demonstrate that a model which incorporates contrarian trading strategies results in more dynamic behaviour in steady state, and hence is able to produce more realistic results. We also demonstrate that a non-learning contrarian model that performs dynamic strategy switching produces long memory phenomena and therefore that learning is not necessary.