Estimating Nonlinear Heterogeneous Agents Models with Neural Networks

Hanno Kase, Leonardo Melosi and Matthias Rottner

No 17391, CEPR Discussion Papers from Centre for Economic Policy Research

Abstract: We leverage recent advances in machine learning to develop an integrated framework that globally solves and estimates models featuring agent heterogeneity and nonlinear constraints, while accounting for aggregate risk. The framework enables the solution and estimation of a frontier quantitative Heterogeneous Agent New Keynesian (HANK) model in its nonlinear form, while allowing for the estimation of parameters governing the steady state. We apply it to U.S. data to estimate a HANK model with a zero lower bound (ZLB) constraint. In the estimated model, the interaction between nonlinearities, such as the ZLB, and wealth heterogeneity emerges as a quantitatively important driver of aggregate volatility.

Keywords: Neural; networks (search for similar items in EconPapers)
JEL-codes: C11 C45 D31 E32 E52 (search for similar items in EconPapers)
Date: 2022-06
References: Add references at CitEc
Citations:

Downloads: (external link)
https://cepr.org/publications/DP17391 (application/pdf)

Related works:
Working Paper: Estimating nonlinear heterogeneous agent models with neural networks (2025)
Working Paper: Estimating Nonlinear Heterogeneous Agent Models with Neural Networks (2024)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:cpr:ceprdp:17391

Ordering information: This working paper can be ordered from
https://cepr.org/publications/DP17391

Access Statistics for this paper

More papers in CEPR Discussion Papers from Centre for Economic Policy Research 33 Great Sutton Street, London EC1V 0DX, UK.
Bibliographic data for series maintained by CEPR ().