Agent-Based Computational Economics (ACE)
Research Area: Financial Markets

Last Updated: 31 January 2009

Site Maintained By: Leigh Tesfatsion Department of Economics Iowa State University Ames, Iowa 50011-1070 http://www.econ.iastate.edu/tesfatsi/ tesfatsi AT iastate.edu Online ACE Course (Self-Study eBook) ACE Website

Table of Contents: Key Issues Introductory Readings General Readings Software, Toolkits, and Demos Resource Sites, Groups, and Individual Researchers

Key Issues

• What makes financial assets/markets special?
• What is the "efficient markets hypothesis (EMH)"?
• How should the market mechanisms governing financial asset trading be modelled?
• How should the learning processes of financial traders be represented?
• How should the "fitness" of financial traders be measured? What drives the co-evolution of these fitnesses over time?
• Do sophisticated (informed) traders necessarily drive "noise traders" from financial markets?
• What are the basic "empirical stylized facts" for financial markets? How well are these stylized facts captured by standard financial models? By agent-based financial models?

Introductory Readings

• W. Brian Arthur, "Complexity in Economic and Financial Markets," Complexity, Vol. 1, No. 1, 1995, pp. 20-25.
  Abstract: This paper provides, among other things, a brief summary of the Santa Fe Artificial Stock Market (SF-ASM) model developed by Arthur et al. (1997) -- see "General Readings," below.

  W. M. van den Bergh, K. Boer, A. de Bruin, U. Kaymak, and J. Spronk, "On Intelligent Agent-Based Analysis of Financial Markets" (pdf,341K), Working Paper, Erasmus University, Rotterdam, 2002.

  J. Doyne Farmer and Andrew W. Lo, "Frontiers of Finance: Evolution and Efficient Markets" (pdf,7pp,105K), SFI Working Paper, April 1999.

  Cars Hommes, "Heterogeneous Agent Models in Economics and Finance", in Leigh Tesfatsion and Kenneth L. Judd (editors), Handbook of Computational Economics, Vol. 2: Agent-Based Computational Economics, Handbooks in Economics Series, North-Holland/Elsevier, Amsterdam, Spring 2006.

  Abstract: This chapter surveys work on dynamic heterogeneous agent models (HAMs) in economics and finance. Emphasis is given to simple models that, at least to some extent, are tractable by analytic methods in combination with computational tools. Most of these models are behavioral models with boundedly rational agents using different heuristics or rule of thumb strategies that may not be perfect, but perform reasonably well. Typically these models are highly nonlinear, e.g. due to evolutionary switching between strategies, and exhibit a wide range of dynamical behavior ranging from a unique stable steady state to complex, chaotic dynamics. Aggregation of simple interactions at the micro level may generate sophisticated structure at the macro level. Simple HAMs can explain important observed stylized facts in financial time series, such as excess volatility, high trading volume, temporary bubbles and trend following, sudden crashes and mean reversion, clustered volatility and fat tails in the returns distribution.

  Blake LeBaron, "Agent-Based Computational Finance", in Leigh Tesfatsion and Kenneth L. Judd (editors), Handbook of Computational Economics, Vol. 2: Agent-Based Computational Economics, Handbooks in Economics Series, North-Holland/Elsevier, Amsterdam, Spring 2006.

  Abstract: This chapter surveys research on agent-based models used in finance. It concentrates on models where the use of computational tools is critical for the process of crafting models that give insights into the importance and dynamics of investor heterogeneity in many financial settings.

  Blake LeBaron, "Agent-Based Computational Finance: Suggested Readings and Early Research" (pdf,150K), Journal of Economic Dynamics and Control 24:5-7 (2000), 679-702. Published article available at Science Direct.

  T. Lux and M. Marchesi (guest editors), Special Issue on "Heterogeneous Interacting Agents in Financial Markets," Journal of Economic Behavior and Organization 49, No. 1, in press. Article available at Science Direct.

  Leigh Tesfatsion, "Introduction to Financial Markets" (html).

  Leigh Tesfatsion, "Information, Bubbles, and the Efficient Markets Hypothesis" (html).

  Edward P. K. Tsang and Serafin Martinez-Jaramillo, Computational Finance (pdf,310K,6pp), Feature Article, IEEE Computational Intelligence Society, August 2004.

  Abstract: This paper briefly outlines the scope and agenda of computational finance research.

General Readings

• George Akerlof and Robert J. Shiller, Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism, Princeton University Press, 2009.
  Note: A short summary of the theme of this book, which provides the essence of George Ackerlof's 2009 T. W. Schultz Memorial Address at the 2009 ASSA meeting in San Francisco in January 2009, can be obtained here (pdf,21K).

  Luca Arciero, Claudia Biancotti, Leandro D'Aurizio and Claudio Impenna, Exploring Agent-Based Methods for the Analysis of Payment Systems: A Crisis Model for StarLogo TNG (html), Journal of Artificial Societies and Social Simulation 12(1)2,2009.

  Abstract: "This paper presents an exploratory agent-based model of a real time gross settlement (RTGS) payment system. Banks are represented as agents who exchange payment requests, which are then settled according to a set of simple rules. The model features the main elements of a real-life system, including a central bank acting as liquidity provider, and a simplified money market. A simulation exercise using synthetic data of BI-REL (the Italian RTGS) predicts the macroscopic impact of a disruptive event on the flow of interbank payments. In our reduced-scale system, three hypothetical distinct phases emerge after the disruptive event: 1) a liquidity sink effect is generated and the participants' liquidity expectations turn out to be excessive; 2) an illusory thickening of the money market follows, along with increased payment delays; and, finally 3) defaulted obligations dramatically rise. The banks cannot staunch the losses accruing on defaults, even after they become fully aware of the critical event, and a scenario emerges in which it might be necessary for the central bank to step in as liquidity provider."

  W. Brian Arthur, John H. Holland, Blake LeBaron, Richard Palmer, and Paul Tayler, "Asset Pricing under Endogenous Expectations in an Artificial Stock Market" (pdf,120K), December 1996 preprint. Final version published as pages 15-44 in W. Brian Arthur, Steven N. Durlauf, and David A. Lane, The Economy as an Evolving Complex System II, Santa Fe Institute Studies in the Sciences of Complexity, Vol. XXVII, Addison-Wesley, 1997.

  Abstract: This paper develops the Santa Fe Artificial Stock Market Model.

  David F. Batten, "Coevolving Markets" (Chapter 7, pages 209-245), in Discovering Artificial Economics, Westview Press, 2000.

  S.-H. Chen and C.-H. Yeh, "Evolving Traders and the Business School with Genetic Programming: A New Architecture of the Agent-Based Stock Market," Journal of Economic Dynamics and Control 25 (3-4), March 2001, pages 363-393. Article available at Science Direct.

  S.-H. Chen, T. Lux, and M. Marchesi, "Testing for Nonlinear Structure in an Artificial Financial Market," Journal of Economic Behavior and Organization 46 (2001), 327-342. Article available at Science Direct.

  John Duffy, "Learning to Speculate: Experiments with Artificial and Real Agents," Journal of Economic Dynamics and Control 25(3/4), March 2001, pages 295-319. Article available at Science Direct.

  Abstract: This paper employs parallel experiments with real and computational agents to explore issues originally raised by Kiyotaki and Wright in their well-known search model of money (JPE, 1989). The primary issue of interest is how individuals might come to accept or learn to adopt a convention in which the particular commodity functioning as "money" is dominated in rate of return by other assets, in the sense that it has a higher storage cost. The key offsetting factor is anticipations ("speculation") concerning the ease with which the "money" good can be turned over in trade for other goods that agents have a higher desire to consume. The author shows how each type of experiment can contribute to the experimental design and interpretation of results for the other.

  J. Doyne Farmer and John Geanakoplos (eds.), Beyond Equilibrium and Efficiency, 352pp., Oxford University Press, 2005. ISBN: 0-195-15094-5

  Abstract: From the publisher: "This book presents recent thought on market efficiency, using a complex systems approach to move past equilibrium models and quantify the actual efficiency of markets. The older view that markets are perfectly efficient has come under attack from several different directions, including studies of market anomalies, human psychology, bounded rationality, agent-based modeling, and evolutionary game theory. This volume brings together some of the best economists, physicists, and biologists working on quantitative models of complex self-organized behavior relevant to measuring market efficiency, to stimulate new approaches to understanding financial markets."

  J. Doyne Farmer is McKinsey Professor at the Santa Fe Institute, and John Geanakoplos is Professor of Economics at Yale University.

  Jens Grossklags, Carsten Schmidt, and Jonathan Siegel, "Dumb Software Agents on an Experimental Asset Market" (pdf,22pp.), Working Paper, School of Information and Management Systems, UC Berkeley.

  Arvid O. I. Hoffmann, Wander Jager, and J. Henk Von Eije, Social Simulation of Stock Markets: Taking It to the Next Level (html), Journal of Artificial Societies and Social Simulation, Vol. 10, No. 2,7, March 31, 2007.

  Abstract: This paper studies the use of social simulation in linking micro level investor behaviour and macro level stock market dynamics. Empirical data from a survey on individual investors' decision-making and social interaction were used to formalize the trading and interaction rules of the agents of an artificial stock market called SimStock Exchange (SSE). Multiple simulation runs were performed with this artificial stock market, which generated macro level results such as stock market prices and returns over time. These outcomes were subsequently compared to empirical macro level data from real stock markets. Partial qualitative and quantitative agreement between the simulated asset returns distributions and the asset returns distributions of the real stock markets were found.

  Two versions of the SSE are available for downloading from the above site, together with a manual: (1) a stand-alone executable version for Windows operating systems; and (2) a Java archive for MacOS and Linux operating systems.

  Paul Johnson, "What I Learned from the Artificial Stock Market" (pdf,20pp,107K), Working Paper, Department of Political Science, University of Kansas, November 5, 2001.

  Abstract: This essay describes some of the changes that were incorporated in the ASM-2.2 revision of the code for the Santa Fe Artificial Stock Market model. It also presents some important lessons for agent-based modelers that can be illustrated with the code.

  Deddy P. Koesrindartoto, "Treasury Auctions, Uniform or Discriminatory?: An Agent-Based Approach" (html), Economics Working Paper No. 04013, Department of Economics, Iowa State University, July 2004.

  Abstract: This study develops an agent-based computational economics (ACE) framework to explore experimentally how a Treasury should auction its securities. Buyers are modeled as profit seekers capable of submitting strategic bids via reinforcement learning. Two distinct auction pricing rules are considered, uniform and discriminatory. The author shows that these two rules result in systematically different auction outcomes under different treatment conditions for relative capacity and for price volatility in a secondary security market. In particular, which auction pricing rule generates greater Treasury revenues varies systematically with these treatment factor specifications. These findings help to explain why previous Treasury auction studies attempting to determine "the" best Treasury auction pricing rule have reached contradictory conclusions.

  Blake LeBaron, "A Builder's Guide to Agent-Based Financial Markets" (pdf,207K), Quantitative Finance 1 (2001), 254-261.

  Blake LeBaron, "Building the Santa Fe Artificial Stock Market" (pdf,123K), Working Paper, Brandeis University, June 2002.

  Abstract: LeBaron provides an insider's look at the construction of the Santa Fe Artificial Stock Market model. He considers the many design questions that went into building the model from the perspective of a decade of experience with agent-based financial markets. He also provides an assessment of the model's overall strengths and weaknesses.

  Blake LeBaron, "Calibrating an Agent-Based Financial Market" (pdf,44pp), Graduate School of International Economics and Finance, Working Paper, Brandeis University, Revised March 2003.

  Abstract: This paper develops an agent-based computational stock market with market participants who adapt and evolve their behaviors over time. The market model is calibrated to match the variability and growth of dividend payments in U.S. data. The market model generates some features that are remarkably similar to those from actual U.S. data, including the volatility of the dividend process, the persistence in volatility and volume, and fat-tailed return distributions.

  Blake LeBaron, W. Brian Arthur, and Richard Palmer, "Time Series Properties of an Artificial Stock Market Model" (pdf,324K), Journal of Economic Dynamics and Control 23 (1999), 1487-1516.

  Abstract: This paper privides a rigorous technical discussion of the Santa Fe Artificial Stock Market Model, including implementation details. Anyone interested in the actual implementation of this model should consult this paper in addition to Arthur et al. (1997) cited above; see also the overview and detailed notes on this model by Tesfatsion below.

  Scott C. Linn and Nicholas S. P. Tay, "Complexity and the Character of Stock Returns: Empirical Evidence and a Model of Asset Prices Based on Complex Investor Learning" (pdf,207K), Management Science, Vol. 53, No. 7, July 2007, pp. 1165-1180.

  Abstract: "We propose a model of complex, self-referential learning and reasoning amongst economic agents that jointly produces security returns consistent with...general observed facts and which are supported here by empirical results presented for a benchmark sample of 50 stocks traded on the New York Stock Exchange. The market we postulate is populated by traders who reason inductively while compressing information into a few fuzzy notions that they can in turn process and analyze with fuzzy logic. We analyze the implications of such behavior for the returns on risky securities within the context of an artificial stock market model. Dynamic simulation experiments of the market are conducted, from which market-clearing prices emerge, allowing us to then compute realized returns. We test the effects of varying values of the parameters of the model on the character of the simulated returns. The results indicate that the model proposed in this paper can jointly account for the presence of a power-law characterization of the positive tail of the survivor function of returns with exponent on the order of three, for autoregressive conditional heteroskedasticity, for long memory in volatility, and for general nonlinear dependencies in returns."

  T. H. Noe, M. J. Rebello, and J. Wang, "Corporate Financing: An Agent-Based Analysis," Journal of Finance, Vol. 58, 943-973, June 2003.

  Nicholas S. P. Tay and Scott C. Linn, "Fuzzy Inductive Reasoning, Expectation Formation, and the Behavior of Security Prices," Journal of Economic Dynamics and Control 25, March 2001, pages 321-361. Article available at Science Direct.

  Leigh Tesfatsion, The Santa Fe Artificial Stock Market: Overview (pdf,69K)

  Leigh Tesfatsion, "Detailed Notes on the Santa Fe Artificial Stock Market Model" (html).

  Frank Westerhoff, "Speculative Markets and the Effectiveness of Price Limits", Journal of Economic Dynamics and Control 28 (2003), 493-508. Article available at Science Direct.

Software, Toolkits, and Demos

• SimStockExchange: Multi-Agent Simulation Stock Market Model (Java, Microsoft Windows Only) (homepage)

• Santa Fe Stock Market Demonstration Software (homepage)
  The Santa Fe Institute (SFI) has open-sourced the Santa Fe Artificial Stock Market (ASM) simulation model, originally developed by a number of SFI researchers in Objective C using the Swarm toolkit. The latest ASM-Swarm file releases can be downloaded from the above ASM sourceforge page and repository maintained by Paul E. Johnson (Political Science, University of Kansas, Lawrence).

  Annotated list of pointers to ACE/CAS General Software and Toolkits

  Annotated list of pointers to ACE/CAS Computational Laboratories and Demonstration Software

  Blake LeBaron (Brandeis University) maintains list of pointers to Interactive Finance Sites that permit users to enter their own information, test hypotheses, watch actual data move across the screen, and perform many other interactive functions.

  The Adaptive Modeler, developed by Jim Witkam (Altreva, Inc.), creates agent-based market simulation models for price forecasting of real world stocks, currencies or other market traded securities. The agent-based model simulates a financial market consisting of thousands of agents whose (technical) trading rules evolve through a special adaptive form of genetic programming. The evolution of trading rules combined with market pricing dynamics drives the agent population to learn to recognize and anticipate recurring price patterns while adapting to changing market behavior. Forecasts can be based on either the behavior of all agents or on a dynamic group of the best performing agents. For ACE researchers this application may be of interest to study the behavior and emergent predictive abilities of an agent-based market model that includes information from a real-world market. Several model initialization options are included such as a user configurable genetic programming engine for the creation of trading rules. Simulation of zero intelligence trading is also supported. Various population statistics and other data can be visualized in charts, distribution histograms and scatter plots, all in real-time. Data can be exported to CSV files for further analysis in other applications. A free (non-expiring) evaluation version with extensive documentation can be downloaded from the Adaptive Modeler homepage, above. Adaptive Modeler is targeted for Windows platforms and requires an installation of Microsoft .Net 2.0 or higher.

Resource Sites, Groups, and Individual Researchers

• Financial Markets and Institutions: Web Resources

• General Resources for Macroeconomics and Financial Economics

• Intensive Course in Financial Instability and Crises, The Ninth Trento Summer School, sponsored by the CEEL Program in Adaptive Economic Dynamics, June/July 2008.

• The CCFEA (Centre for Computational Finance and Economic Agents, University of Essex, UK) is an interdisciplinary research centre. Fields of research include computational finance, evolutionary methods for finance and economics, simulations of artificially intelligent agents in markets, market and policy design. CCFEA offers programmes leading to MSc and PhD in Computational Finance, Computational Economics and in Financial Software Engineering. Selected course materials can be found here.

• Shu-Heng Chen (Economics, National Chengchi University, Taipei, Taiwan): Incorporating learning into economic models; Genetic programming models of learning; Financial market modelling; Auctions.

• J. Doyne Farmer (Santa Fe Institute, New Mexico): Evolution and the efficiency of financial markets; Finance.

• Christophre Georges (Economics, Hamilton College, Clinton, NY): Learning and agent interactions in macroeconomics and finance; Disequilibrium dynamics.

• Cars Hommes (Center for Nonlinear Dynamics in Economics and Finance - CeNDEF, University of Amsterdam, The Netherlands): Complex adaptive systems; Multi-agent systems; Evolutionary dynamics; Expectations and learning; Bounded rationality; Bifurcations and chaos.

• Blake LeBaron (Graduate School of International Economics and Finance, Brandeis University, Waltham, Massachusetts): Quantitative dynamics of interacting systems of adaptive agents, and how these systems replicate real world phenomena; Behavior of traders in financial markets; Nonlinear behavior of financial and macroeconomic time series. LeBaron maintains a Web site on Agent-Based Computational Finance (ACF). ACF is an application of agent-based computational methods to finance and financial markets. The Web site is designed to give researchers interested in this area a starting point in terms of finding relevant online materials. Resources incorporated to date include pointers and paper lists. LeBaron also maintains a list of pointers to Interactive Finance Sites that permit users to enter their own information, test hypotheses, watch actual data move across the screen, and perform many other interactive functions.

• Scott C. Linn (Division of Finance, University of Oklahoma, Norman, OK): Corporate finance; Corporate governance; Behaviors of security prices and energy markets; Artificial stock market modeling with fuzzy inductive reasoning.

• Thomas Lux (Department of Economics, University of Kiel, Germany): Agent-based computational finance; Stock market dynamics; Speculative bubbles.

• Frank H. Westerhoff (Department of Economics, University of Osnabrueck, Germany): Financial market dynamics; Technical and fundamental trading rules; Nonlinear dynamics and chaos; Bounded rationality and behavioral finance; Regulation/design of financial markets.

Copyright © 2008 Leigh Tesfatsion. All Rights Reserved.