Template-Type: ReDIF-Paper 1.0
Author-Name: Niko Hauzenberger
Author-Name-First: Niko
Author-Name-Last: Hauzenberger
Author-Workplace-Name: University of Salzburg
Author-Name: Florian Huber
Author-Name-First: Florian
Author-Name-Last: Huber
Author-Workplace-Name: University of Salzburg
Author-Name: Gary Koop
Author-Name-First: Gary
Author-Name-Last: Koop
Author-Workplace-Name: University of Strathclyde
Author-Name: James Mitchell
Author-Name-First: James
Author-Name-Last: Mitchell
Author-Workplace-Name: Federal Reserve Bank of Cleveland
Title: Bayesian Modelling of TVP-VARs Using Regression Trees
Abstract: In light of widespread evidence of parameter instability in
macroeconomic models, many time-varying parameter (TVP) models have
been proposed. This paper proposes a nonparametric TVP-VAR model
using Bayesian additive regression trees (BART) that models the TVPs as
an unknown function of effect modifiers. The novelty of this model arises
from the fact that the law of motion driving the parameters is treated
nonparametrically. This leads to great flexibility in the nature and extent
of parameter change, both in the conditional mean and in the conditional
variance. Parsimony is achieved through adopting nonparametric factor
structures and use of shrinkage priors. In an application to US macroeconomic
data, we illustrate the use of our model in tracking both the evolving
nature of the Phillips curve and how the effects of business cycle shocks on
in inflation measures vary nonlinearly with changes in the effect modifiers.
Length:  pages
Creation-Date: 2020-02
Revision-Date: 2023-08
Publication-Status: 
File-URL: https://www.strath.ac.uk/media/1newwebsite/departmentsubject/economics/research/researchdiscussionpapers/2023/23-08.pdf
File-Format: Application/pdf
Number: 2308
Classification-JEL: C11, C32, C51, E31, E32
Keywords: Bayesian vector autoregression; Time-varying parameters; Nonparametric modeling; Machine learning; Regression trees; Phillips curve; Business cycle shocks
Handle: RePEc:str:wpaper:2308