A Generic Architecture for Realistic Simulations of Complex Financial Dynamics

Artificial Stock Markets (here after ASM) have received an increasing amount of academic interest since the seminal works of [20] or [17]. Such platforms have benefitted from advances and new methods developed in the field of multi agent-systems (see for example [10], [14] and [27]). These Agent-Based virtual environments are particularly useful to study various aspects of the financial world in an entirely controlled environment, opening new perspectives for policy makers, regulatory institutions and firms developing business solutions in the financial industry (for example asset management or trading). There is little doubt ASM could become a key system in the post financial crisis risk-management toolbox to overcome the weaknesses of traditional approaches. Agents-based modeling and simulation offer frameworks to study the impact of a Tobin's tax for example, or to develop new stress tests for assessing financial resilience to economic shocks or to develop new automatic trading techniques. In this research paper, we introduce a new, highly flexible Agent-Based model of financial markets in an API form. This application offers a solution for implementing realistic simulations of complex financial dynamics using both artificial intelligence, distributed agents and realistic market algorithms. We consider that if various questions in Finance can be solved with Agent-Based Modeling solutions, Multi-Agent Systems directly benefit from Financial questions as well : ASMs, with driving simulators (see [9]), offer one of the richest environments to evolve software engineering for multi-agent systems. Therefore, the punchline of this paper is that development of artificial markets offers the whole variety of issues one can face in agent based mod-eling. Among others, ASM are grounded on an individual-based approach with local interaction, distributed knowledge and resources, heterogeneous environments, agent autonomy, artificial intelligence, speech acts, discrete scheduling and simulation. We defend this punchline with an illustration derived from the building of the ArTificial Open Market API (hereafter ATOM, see http://atom.univ-lille1.fr). This paper is organized as follows. In a first section we briefly present the relevant literature around artificial markets and Agent-Based modeling in Finance. In a second section, we extensively introduce the ArTificial Open Market API and present the mains computing issues linked to its development. In the third section, we present agent's behavior and introduce a series of tests geared at verifying the efficiency of our API. We then conclude and open new research and technological perspectives.