Dynamic Mixture Vector Autoregressions with Score-Driven Weights

37 Pages Posted: 16 Feb 2022

See all articles by Alexander Georges Gretener

Alexander Georges Gretener

University of Kiel - Institute for Quantitative Business and Economic Research (QBER)

Matthias Neuenkirch

University of Trier - Faculty of Economics; CESifo (Center for Economic Studies and Ifo Institute)

Dennis Umlandt

University of Innsbruck - Department of Banking and Finance

Multiple version iconThere are 2 versions of this paper

Date Written: February 15, 2022

Abstract

We propose a novel dynamic mixture vector autoregressive (VAR) model in which time-varying mixture weights are driven by the predictive likelihood score. Intuitively, the state weight of the k-th component VAR model in the subsequent period is increased if the current observation is more likely to be drawn from this particular state. The model is not limited to a specific distributional assumption and allows for straightforward likelihood-based estimation and inference. We conduct a Monte Carlo study and find that the score-driven mixture VAR model is able to adequately filter the mixture dynamics from a variety of different data generating processes which most other observation-driven dynamic mixture VAR models cannot appropriately cope with. Finally, we illustrate our approach by an application where we model the conditional joint distribution of economic and financial conditions and derive generalized impulse responses.

Keywords: Dynamic Mixture Models, Generalized Autoregressive Score Models, Macro-Financial Linkages, Nonlinear VAR

JEL Classification: C32, C34, G17

Suggested Citation

Gretener, Alexander Georges and Neuenkirch, Matthias and Umlandt, Dennis, Dynamic Mixture Vector Autoregressions with Score-Driven Weights (February 15, 2022). Available at SSRN: https://ssrn.com/abstract=4035650 or http://dx.doi.org/10.2139/ssrn.4035650

Alexander Georges Gretener

University of Kiel - Institute for Quantitative Business and Economic Research (QBER) ( email )

Heinrich-Hecht-Platz 9
Kiel, 24106
Germany
+49 (0) 431 / 880 - 5598 (Phone)

HOME PAGE: http://www.qber.uni-kiel.de

Matthias Neuenkirch (Contact Author)

University of Trier - Faculty of Economics ( email )

Universitätsring 15
Trier, 54296
Germany
+49 - (0)651 - 201 - 2629 (Phone)

HOME PAGE: http://www.uni-trier.de/index.php?id=50130

CESifo (Center for Economic Studies and Ifo Institute) ( email )

Poschinger Str. 5
Munich, DE-81679
Germany

Dennis Umlandt

University of Innsbruck - Department of Banking and Finance ( email )

Innsbruck
Austria

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
69
Abstract Views
316
Rank
479,539
PlumX Metrics