Quotation Huber, Florian, Kastner, Gregor, Feldkircher, Martin. 2019. Should I stay or should I go? A latent threshold approach to large-scale mixture innovation models. Journal of Applied Econometrics. 34 (5), 621-640.




We propose a straightforward algorithm to estimate large Bayesian time-varying parameter vector autoregressions with mixture innovation components for each coefficient in the system. The computational burden becomes manageable by approximating the mixture indicators driving the time-variation in the coefficients with a latent threshold process that depends on the absolute size of the shocks. Two applications illustrate the merits of our approach. First, we forecast the US term structure of interest rates and demonstrate forecast gains relative to benchmark models. Second, we apply our approach to US macroeconomic data and find significant evidence for time-varying effects of a monetary policy tightening.


Press 'enter' for creating the tag

Publication's profile

Status of publication Published
Affiliation WU
Type of publication Journal article
Journal Journal of Applied Econometrics
Citation Index SSCI
WU Journalrating 2009 A
WU-Journal-Rating new FIN-A, VW-B, WH-B
Language English
Title Should I stay or should I go? A latent threshold approach to large-scale mixture innovation models
Volume 34
Number 5
Year 2019
Page from 621
Page to 640
Reviewed? Y
URL https://onlinelibrary.wiley.com/doi/full/10.1002/jae.2680
DOI https://doi.org/10.1002/jae.2680
Open Access Y
Open Access Link https://onlinelibrary.wiley.com/doi/full/10.1002/jae.2680
JEL C11, C32, C50, E43, E52


Kastner, Gregor (Details)
Feldkircher, Martin (Oesterreichische Nationalbank (OeNB), Austria)
Huber, Florian (University of Salzburg, Austria)
Institute for Statistics and Mathematics IN (Details)
Research areas (Ă–STAT Classification 'Statistik Austria')
1105 Computer software (Details)
1162 Statistics (Details)
5323 Econometrics (Details)
5701 Applied statistics (Details)
5707 Time series analysis (Details)
Google Scholar: Search