We propose two novel methods to “bring Agent Based Models (ABMs) to the data”. First, we describe a Bayesian procedure to estimate the numerical values of ABM parameters that takes into account the time structure of simulated and observed time series. Second, we propose a method to forecast aggregate time series using data obtained from the simulation of an ABM. We apply our methodological contributions to a specific medium-scale macro ABM.
|Number of pages||28|
|Journal||JOURNAL OF ECONOMIC BEHAVIOR & ORGANIZATION|
|Publication status||Published - 2020|
- Agent Based Models