Understanding the workings (economic performance, stability, etc.) of today's financial systems requires consideration of critical
computational ingredients in their behavior. Agent-based modeling (ABM) affords direct representation of these computational elements,and accommodates the heterogeneity and complexity characteristic of financial environments. Although ABM is typically presented as an alternative to mainstream economic approaches, I argue that it is actually compatible and even complementary with standard frameworks, and demonstrate this with two recent finance-related studies. The first concerns the functioning of information and trust networks in the systemic flow of credit. Using empirical game-theoretic methods, we investigate conditions for the formation of viable credit networks by self-interested agents. The second investigates implications of the trend toward trading in securities by algorithmic entities, at unprecedented degrees of speed and autonomy. Through an agent-based study of high-frequency trading on fragmented markets, we find that latency arbitrage degrades allocative efficiency. The results of this study also support our proposal to switch from continuous-time trading to a discrete-time mechanism, using one-second call markets.
Michael P. Wellman is Professor of Computer Science & Engineering at the University of Michigan. He received a PhD from the
Massachusetts Institute of Technology in 1988 for his work in qualitative probabilistic reasoning and decision-theoretic planning.
From 1988 to 1992, Wellman conducted research in these areas at the USAF’s Wright Laboratory. For the past 20+ years, his research has focused on computational market mechanisms for distributed decision making and electronic commerce. As Chief Market Technologist for TradingDynamics, Inc. (now part of Ariba), he designed configurable auction technology for dynamic business-to-business commerce. Wellman previously served as Chair of the ACM Special Interest Group on Electronic Commerce (SIGecom), and as Executive Editor of the Journal of Artificial Intelligence Research. He is a Fellow of the Association for the Advancement of Artificial Intelligence and the Association for Computing Machinery.
This talk is partially sponsored by an NSF grant to develop a "data science for finance" research platform. Please contact Louiqa Raschid louiqa@umiacs.umd.edu if you wish to meet with the speaker.A version of this talk will be presented earlier in the day at the
Office of Financial Research, Department of the Treasury. Please contact Mark Flood mark.flood@treasury.gov for details.