A Comparison of Economic Agent-Based Model Calibration Methods

Interest in agent-based models of financial markets and the wider economy has increased consistently over the last few decades, in no small part due to their ability to reproduce a number of empirically-observed stylised facts that are not easily recovered by more traditional modelling approaches. Nevertheless, the agent-based modelling paradigm faces mounting criticism, focused particularly on the rigour of current validation and calibration practices, most of which remain qualitative and stylised fact-driven. While the literature on quantitative and data-driven approaches has seen significant expansion in recent years, most studies have focused on the introduction of new calibration methods that are neither benchmarked against existing alternatives nor rigorously tested in terms of the quality of the estimates they produce. We therefore compare a number of prominent ABM calibration methods, both established and novel, through a series of computational experiments in an attempt to determine the respective strengths and weaknesses of each approach and the overall quality of the resultant parameter estimates. We find that Bayesian estimation, though less popular in the literature, consistently outperforms frequentist, objective function-based approaches and results in reasonable parameter estimates in many contexts. Despite this, we also find that agent-based model calibration techniques require further development in order to definitively calibrate large-scale models. We therefore make suggestions for future research.

[1]  Mattia Guerini,et al.  A Method for Agent-Based Models Validation , 2016 .

[2]  Sylvain Barde A Practical, Accurate, Information Criterion for Nth Order Markov Processes , 2017 .

[3]  Takao Terano,et al.  Meeting the Challenge of Social Problems via Agent-Based Simulation , 2003 .

[4]  Sylvain Barde,et al.  Direct comparison of agent-based models of herding in financial markets , 2016 .

[5]  R. Franke Applying the method of simulated moments to estimate a small agent-based asset pricing model☆ , 2009 .

[6]  T. Lux,et al.  Estimation of Sentiment Effects in Financial Markets: A Simulated Method of Moments Approach , 2018 .

[7]  Ali Kaveh,et al.  Advances in Metaheuristic Algorithms for Optimal Design of Structures , 2014 .

[8]  Jakob Grazzini,et al.  Analysis of the Emergent Properties: Stationarity and Ergodicity , 2012, J. Artif. Soc. Soc. Simul..

[9]  Peter Winker,et al.  An objective function for simulation based inference on exchange rate data , 2007 .

[10]  陳樹衡,et al.  Agent-Based Computational Macroeconomics: A Survey , 2002 .

[11]  John Geanakoplos,et al.  The virtues and vices of equilibrium and the future of financial economics , 2009 .

[12]  Jozef Barunik,et al.  Estimation of Financial Agent-Based Models with Simulated Maximum Likelihood , 2016 .

[13]  Y. Shtarkov,et al.  The context-tree weighting method: basic properties , 1995, IEEE Trans. Inf. Theory.

[14]  Francesco Lamperti,et al.  An Information Theoretic Criterion for Empirical Validation of Simulation Models , 2018 .

[15]  G. Dosi,et al.  Schumpeter meeting Keynes: A policy-friendly model of endogenous growth and business cycles , 2010 .

[16]  Isabelle Salle,et al.  Efficient Sampling and Metamodeling for Computational Economic Models , 2012 .

[17]  F. Westerhoff,et al.  Structural stochastic volatility in asset pricing dynamics: Estimation and model contest , 2012 .

[18]  Matteo G. Richiardi,et al.  Estimation of ergodic agent-based models by simulated minimum distance , 2015 .

[19]  Michael J. North,et al.  Tutorial on agent-based modelling and simulation , 2005, Proceedings of the Winter Simulation Conference, 2005..

[20]  Kevin P. Murphy,et al.  Machine learning - a probabilistic perspective , 2012, Adaptive computation and machine learning series.

[21]  M. Raberto,et al.  Credit Money and Macroeconomic Instability in the Agent-based Model and Simulator Eurace , 2010 .

[22]  Christine A. Shoemaker,et al.  Constrained Global Optimization of Expensive Black Box Functions Using Radial Basis Functions , 2005, J. Glob. Optim..

[23]  G. Fagiolo,et al.  Rock around the clock: An agent-based model of low- and high-frequency trading , 2014, Journal of Evolutionary Economics.

[24]  W. Brock,et al.  Heterogeneous beliefs and routes to chaos in a simple asset pricing model , 1998 .

[25]  Thomas Lux,et al.  Empirical validation of agent-based models , 2018 .

[26]  Mauro Gallegati,et al.  Validating and Calibrating Agent-Based Models: A Case Study , 2007 .

[27]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

[28]  Thomas Lux,et al.  Estimation of a simple agent-based model of financial markets: An application to Australian stock and foreign exchange data , 2006 .

[29]  Giorgio Fagiolo,et al.  Validation of Agent-Based Models in Economics and Finance , 2019, Simulation Foundations, Methods and Applications.

[30]  Amir Sani,et al.  Agent-Based Model Calibration Using Machine Learning Surrogates , 2017, 1703.10639.

[31]  Paul Knysh,et al.  Blackbox: A procedure for parallel optimization of expensive black-box functions , 2016, ArXiv.

[32]  Sylvain Barde,et al.  An Empirical Validation Protocol for Large-Scale Agent-Based Models , 2017 .

[33]  Thomas Lux,et al.  Empirical validation of stochastic models of interacting agents , 2007 .

[34]  Annalisa Fabretti,et al.  On the problem of calibrating an agent based model for financial markets , 2012, Journal of Economic Interaction and Coordination.

[35]  D. Foley,et al.  The economy needs agent-based modelling , 2009, Nature.

[36]  Giovanni Dosi,et al.  On the robustness of the fat-tailed distribution of firm growth rates: a global sensitivity analysis , 2016 .

[37]  R. Cont Empirical properties of asset returns: stylized facts and statistical issues , 2001 .

[38]  James J. Heckman,et al.  The Empirical Foundations of Calibration , 1996 .

[39]  D. McFadden A Method of Simulated Moments for Estimation of Discrete Response Models Without Numerical Integration , 1989 .

[40]  J. Farmer,et al.  The price dynamics of common trading strategies , 2000, cond-mat/0012419.

[41]  Peter Winker,et al.  A global optimization heuristic for estimating agent based models , 2003, Comput. Stat. Data Anal..

[42]  J. Farmer,et al.  Macroprudential Policy in an Agent-Based Model of the UK Housing Market , 2016, SSRN Electronic Journal.

[43]  Tim Gebbie,et al.  Can agent-based models probe market microstructure? , 2016, Physica A: Statistical Mechanics and its Applications.

[44]  F. Lamperti An Information Theoretic Criterion for Empirical Validation of Time Series Models , 2015 .

[45]  B. LeBaron Agent-based Computational Finance , 2006 .

[46]  Giorgio Fagiolo,et al.  Macroeconomic Policy in DSGE and Agent-Based Models Redux: New Developments and Challenges Ahead , 2016 .

[47]  Matteo G. Richiardi,et al.  Bayesian Estimation of Agent-Based Models , 2017 .

[48]  Robert L. Axtell,et al.  An agent-based model of the housing market bubble in metropolitan Washington, D.C. , 2014 .

[49]  T. Lux,et al.  Estimation of Agent-Based Models: The Case of an Asymmetric Herding Model , 2005 .

[50]  Mark Harman,et al.  Agent-Based Modelling of Stock Markets Using Existing Order Book Data , 2012, MABS.