Black-box Bayesian inference for economic agent-based models

Abstract

Simulation models, in particular agent-based models, are gaining popularity in economics. The considerable flexibility they offer, as well as their capacity to reproduce a variety of empirically observed behaviors of complex systems, give them broad appeal, and the increasing availability of cheap computing power has made their use feasible. Yet a widespread adoption in real-world modelling and decision-making scenarios has been hindered by the difficulty of performing parameter estimation for such models. In general, simulation models lack a tractable likelihood function, which precludes a straightforward application of standard statistical inference techniques. A number of recent works (Grazzini et al., 2017;Platt, 2020Platt, , 2021 have sought to address this problem through the application of likelihood-free inference techniques, in which parameter estimates are determined by performing some form of comparison between the observed data and simulation output. However, these approaches are (a) founded on restrictive assumptions, and/or (b) typically require many hundreds of thousands of simulations. These qualities make them unsuitable for large-scale simulations in economics and can cast doubt on the validity of these inference methods in such scenarios. In this paper, we investigate the efficacy of two classes of simulation-efficient black-box approximate Bayesian inference methods that have recently drawn significant attention within the probabilistic machine learning community: neural posterior estimation and neural density ratio estimation. We present a number of benchmarking experiments in which we demonstrate that neural network based black-box methods provide state of the art parameter inference for economic simulation models, and crucially are compatible with generic multivariate time-series data. In addition, we suggest appropriate assessment criteria for use in future benchmarking of approximate Bayesian inference procedures for economic simulation models.

Inference on a Multi-Patch Epidemic Model with Partial Mobility, Residency, and Demography: Case of the 2020 COVID-19 Outbreak in Hermosillo, Mexico

“…To address this, one option for small spatial scales and patch sizes is to leverage computational techniques such as agent-based modeling. This approach captures the important variability that arises from individual interactions [136][137][138][139][140][141]. Additionally, in order to reduce the variability of simulations, relevant contact patterns produced by street or building layouts can be incorporated.…”

Deep Calibration of Market Simulations using Neural Density Estimators and Embedding Networks

“…These methods include classical methods such as approximate Bayesian computation (ABC) as well as more recent advances that leverage deep neural networks such as neural posterior estimation (NPE). For a review of applications of simulation-based inference in the context of economics and financial timeseries, see [17], which demonstrates how these methods can be applied to models of market dynamics. While these methods have been applied in many fields including epidemiology, high-energy physics, and nonequilibrium systems, they have yet to be used to calibrate a market simulator to real market data [5,22,39].…”

Black-it: A Ready-to-Use and Easy-to-Extend Calibration Kit for Agent-based Models

“…In addition, our tool implements a number of standard loss functions, such as the "method of moments" distance (Franke, 2009) and the GSL-divergence (Lamperti, 2018). Currently, only likelihood-free losses are implemented, since probabilistic methods are typically too expensive to be used in large-scale ABMs (Platt, 2020), but nothing hinders the inclusion of the latest advancements in the field (Dyer et al, 2022;Platt, 2021) into our tool.…”