Munich Personal RePEc Archive

Time-Varying Vector Autoregressions: Efficient Estimation, Random Inertia and Random Mean

Legrand, Romain (2019): Time-Varying Vector Autoregressions: Efficient Estimation, Random Inertia and Random Mean.

This is the latest version of this item.


Download (2MB) | Preview
[img] PDF

Download (4MB)


Time-varying VAR models represent fundamental tools for the anticipation and analysis of economic crises. Yet they remain subject to a number of limitations. The conventional random walk assumption used for the dynamic parameters appears excessively restrictive, and the existing estimation procedures are largely inefficient. This paper improves on the existing methodologies in four directions: i) it introduces a general time-varying VAR model which relaxes the standard random walk assumption and defines the dynamic parameters as general autoregressive processes with equation-specific mean values and autoregressive coefficients. ii) it develops an efficient estimation algorithm for the model which proceeds equation by equation and combines the traditional Kalman filter approach with the recent precision sampler methodology. iii) it develops extensions to estimate endogenously the mean values and autoregressive coefficients associated with each dynamic process. iv) through a case study of the Great Recession in four major economies (Canada, the Euro Area, Japan and the United States), it establishes that forecast accuracy can be significantly improved by using the proposed general time-varying model and its extensions in place of the traditional random walk specification.

Available Versions of this Item

Atom RSS 1.0 RSS 2.0

Contact us: mpra@ub.uni-muenchen.de

This repository has been built using EPrints software.

MPRA is a RePEc service hosted by Logo of the University Library LMU Munich.