Template-type: ReDIF-Paper 1.0
Author-Name: Jakob Grazzini
Author-Workplace-Name: Catholic University of Milan, Italy
Author-Name: Matteo Richiardi
Author-Workplace-Name: Institute for New Economic Thinking, Nuffield College and University of Torino
Author-Name: Mike Tsionas
Author-Workplace-Name: Lancaster Univesity Mangament School
Title: Bayesian Estimation of Agent-Based Models


Abstract: We consider Bayesian inference techniques for Agent-Based (AB) models, as an alternative
to simulated minimum distance (SMD). We discuss the specificities of AB models with
respect to models with exact aggregation results (as DSGE models), and how this impact
estimation. Three computationally heavy steps are involved: (i) simulating the model, (ii)
estimating the likelihood and (iii) sampling from the posterior distribution of the parameters.
Computational complexity of AB models implies that efficient techniques have to be
used with respect to points (ii) and (iii), possibly involving approximations. We first discuss
non-parametric (kernel density) estimation of the likelihood, coupled with Markov chain
Monte Carlo sampling schemes. We then turn to parametric approximations of the likelihood,
which can be derived by observing the distribution of the simulation outcomes around
the statistical equilibria, or by assuming a specific form for the distribution of external deviations
in the data. Finally, we introduce Approximate Bayesian Computation techniques for
likelihood-free estimation. These allow embedding SMD methods in a Bayesian framework,
and are particularly suited when robust estimation is needed. These techniques are tested,
for the sake of comparison, in the same price discovery model used by Grazzini and Richiardi
(2015) to illustrate SMD techniques.


X-Classification-JEL: 
X-Keywords:  
Length: 29 pages
Creation-Date: 2015-11-27
Number: 2015-W12
File-URL: https://www.nuffield.ox.ac.uk/economics/papers/2015/AB-v26.pdf
File-Format: application/pdf
Handle: RePEc:nuf:econwp:1512