Predictive density combination using Bayesian machine learning

Tools

Chernis, Tony and Hauzenberger, Niko and Huber, Florian and Koop, Gary and Mitchell, James (2025) Predictive density combination using Bayesian machine learning. International Economic Review. ISSN 0020-6598 (https://doi.org/10.1111/iere.12759)

[thumbnail of Chernis-etal-IER-2025-Predictive-density-combination-using-Bayesian-machine-learning]
Preview
 Text. Filename: Chernis-etal-IER-2025-Predictive-density-combination-using-Bayesian-machine-learning.pdf
Final Published Version
License: Creative Commons Attribution 4.0 logo
Download (673kB)| Preview

Abstract

Bayesian predictive synthesis (BPS) provides a method for combining multiple predictive distributions based on agent/expert opinion analysis theory and encompasses a range of existing density forecast pooling methods. The key ingredient in BPS is a “synthesis” function. This is typically specified parametrically as a dynamic linear regression. In this paper, we develop a nonparametric treatment of the synthesis function using regression trees. We show the advantages of our tree-based approach in two macroeconomic forecasting applications. The first uses density forecasts for GDP growth from the euro area’s Survey of Professional Forecasters. The second combines density forecasts of US inflation produced by many regression models involving different predictors. Both applications demonstrate the benefits – in terms of improved forecast accuracy and interpretability – of modeling the synthesis function nonparametrically.

ORCID iDs

Chernis, Tony, Hauzenberger, Niko ORCID logoORCID: https://orcid.org/0000-0002-2683-8421, Huber, Florian, Koop, Gary ORCID logoORCID: https://orcid.org/0000-0002-6091-378X and Mitchell, James;
CORE (COnnecting REpositories)