Tail forecasting with multivariate Bayesian additive regression trees
Clark, Todd E. and Huber, Florian and Koop, Gary and Marcellino, Massimiliano and Pfarrhofer, Michael (2023) Tail forecasting with multivariate Bayesian additive regression trees. International Economic Review, 64 (3). pp. 979-1022. ISSN 0020-6598 (https://doi.org/10.1111/iere.12619)
Preview |
Text.
Filename: Clark-etal-IER-2023-Tail-forecasting-with-multivariate-Bayesian-additive-regression-trees.pdf
Final Published Version License: Download (5MB)| Preview |
Abstract
We develop multivariate time-series models using Bayesian additive regression trees that posit nonlinearities among macroeconomic variables, their lags, and possibly their lagged errors. The error variances can be stable, feature stochastic volatility, or follow a nonparametric specification. We evaluate density and tail forecast performance for a set of U.S. macroeconomic and financial indicators. Our results suggest that the proposed models improve forecast accuracy both overall and in the tails. Another finding is that when allowing for nonlinearities in the conditional mean, heteroskedasticity becomes less important. A scenario analysis reveals nonlinear relations between predictive distributions and financial conditions.
ORCID iDs
Clark, Todd E., Huber, Florian, Koop, Gary ORCID: https://orcid.org/0000-0002-6091-378X, Marcellino, Massimiliano and Pfarrhofer, Michael;-
-
Item type: Article ID code: 83245 Dates: DateEvent7 August 2023Published6 January 2023Published Online16 November 2022Accepted5 October 2021SubmittedSubjects: Social Sciences > Economic Theory > Methodology > Mathematical economics. Quantitative methods > Econometrics Department: Strathclyde Business School > Economics Depositing user: Pure Administrator Date deposited: 17 Nov 2022 10:01 Last modified: 14 Sep 2024 02:12 URI: https://strathprints.strath.ac.uk/id/eprint/83245