Bayesian Neural Networks for Macroeconomic Analysis

Niko Hauzenberger , Florian Huber, Karin Klieber and Massimiliano Marcellino

No 19381, CEPR Discussion Papers from C.E.P.R. Discussion Papers

Abstract: Macroeconomic data is characterized by a limited number of observations (small T), many time series (big K) but also by featuring temporal dependence. Neural networks, by contrast, are designed for datasets with millions of observations and covariates. In this paper, we develop Bayesian neural networks (BNNs) that are well-suited for handling datasets commonly used for macroeconomic analysis in policy institutions. Our approach avoids extensive specification searches through a novel mixture specification for the activation function that appropriately selects the form of nonlinearities. Shrinkage priors are used to prune the network and force irrelevant neurons to zero. To cope with heteroskedasticity, the BNN is augmented with a stochastic volatility model for the error term. We illustrate how the model can be used in a policy institution through simulations and by showing that BNNs produce more accurate point and density forecasts compared to other machine learning methods.

Keywords: Bayesian neural networks; Model selection; Shrinkage priors; Macro forecasting (search for similar items in EconPapers)
JEL-codes: C11 C30 C45 C53 E3 E44 (search for similar items in EconPapers)
Date: 2024-08
References: Add references at CitEc
Citations:

Downloads: (external link)
https://cepr.org/publications/DP19381 (application/pdf)
CEPR Discussion Papers are free to download for our researchers, subscribers and members. If you fall into one of these categories but have trouble downloading our papers, please contact us at subscribers@cepr.org

Related works:
Working Paper: Bayesian Neural Networks for Macroeconomic Analysis (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:19381

Ordering information: This working paper can be ordered from
https://cepr.org/publications/DP19381

Access Statistics for this paper

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