Economic Forecasting with an Agent-Based Model

Poledna, Sebastian; Miess, MichaelORCID: and Hommes, Cars H. (2020) Economic Forecasting with an Agent-Based Model. IIASA, Advanced Systems Analysis (ASA), Risk & Resilience (RISK), IIASA Working Paper 20-001 DOI: 10.2139/ssrn.3484768

poledna-miess-hommes-2020-forecasting-with-agent-based-model-iiasa-wp.pdf - Published Version
Available under License Creative Commons Attribution Non-commercial.

Download (803kB) | Preview


We develop the first agent-based model (ABM) that can compete with and in the long run significantly outperform benchmark VAR and DSGE models in out-of-sample forecasting of macro variables. Our ABM for a small open economy uses micro and macro data from national sector accounts, input-output tables, government statistics, and census data. The model incorporates all economic activities as classified by the European System of Accounts as heterogeneous agents. The detailed structure of the ABM allows for a breakdown into sector level forecasts. Potential applications of the model include stress-testing and predicting the effects of monetary or fiscal macroeconomic policies.

Item Type: Discussion/ Working Paper (Unspecified)
Keywords: agent-based models, behavioral macro, macroeconomic forecasting, micro data
Classification Codes (e.g. JEL): E70, E32, E37
Research Units: Former Research Units (until 2020) > Macroeconomics and Economic Policy
Related URLs:
Date Deposited: 24 Apr 2020 10:19
Last Modified: 24 Apr 2020 10:19
DOI: 10.2139/ssrn.3484768
ISSN: 1556-5068

Actions (login required)

View Item View Item