Научный архив: статьи

Since the 1980s asset pricing in the traditional neoclassical paradigm has been confronting empirical evidence contradicting both the predictions of the models as well as their microfoundations. Simultaneously, the market microstructure literature started probing the details of the trading process, turning the spotlight onto the effects of asymmetric information, clearing mechanisms and agents’ learning and belief formation. These details, which were “abstracted away” in the earlier models, are becoming ever more important as the complexity of markets grows due to proliferation of algorithmic and high frequency trading and markets turn into ecologies of strategic, but not necessarily perfectly rational, co-evolving agents. In this review article I argue that the paradigms of agent-based and evolutionary finance are ideally suited to handle the modelling of markets as these complex ecologies. I review the most prominent contributions of evolutionary and agent-based modelling to asset pricing, specifically, categorizing them into three main streams: the research on the effects of institutional details of the markets, the research on the effects of agent heterogeneity, and the research of market selection. Furthermore, I argue that further progress can be made by combining the evolutionary and agent-based paradigms and highlight research questions for which such a mixed-method approach is likely to be the most fruitful.