Validation of Agent-Based Models in Economics and Finance

Since the influential survey by Windrum et al. (2007), research on empirical validation of agent-based models in economics has made substantial advances, thanks to a constant flow of high-quality contributions. This Chapter attempts to take stock of such recent literature to offer an updated critical review of existing validation techniques. We sketch a simple theoretical framework that conceptualizes existing validation approaches, which we discuss along three different dimensions: (i) comparison between artificial and real-world data; (ii) calibration and estimation of model parameters; and (iii) parameter space exploration.

working_paper_2017_25_cover

Validation of Agent-Based Models in Economics and Finance

Giorgio Fagiolo
Institute of Economics, Scuola Superiore Sant’Anna

Mattia Guerini
Institute of Economics, Scuola Superiore Sant’Anna and OFCE-Science Po

Francesco Lamperti
Institute of Economics, Scuola Superiore Sant’Anna and FEEM

Alessio Moneta
Institute of Economics, Scuola Superiore Sant’Anna

Andrea Roventini
Institute of Economics, Scuola Superiore Sant’Anna and OFCE-Science Po

Working Paper
25/2017 October