Pandemic Priors BVAR

Pandemic Priors: a simple, easy, and flexible way of estimating Bayesian VARs taking into consideration the pandemic period, as a Minnesota prior with time dummies

Update (Aug 2025): 

• New fully revised version of the paper, including period-specific optimal priors and a DSGE version.  Link here!

• New more efficient MATLAB, Julia, Python, and R versions and a DSGE implementation (here). 

Update (Feb 2024): 

• Added Pandemic Priors extensions of the Giannone, Lenza, and Primiceri (2015) optimal priors (here) and the Chan (2022) asymmetric conjugate priors (here). 

Update (Feb 2023): Inclusion of a test for the optimal level of shrinkage for the pandemic period and a test for suitability of the Pandemic Priors.

Update (Nov 2022): Added the flexibility of how much signal to take from pandemic period.

Paper: Pandemic Priors

Abstract:

A few extreme pandemic observations can distort the estimated persistence of macroeconomic variables, contaminating forecasts and the interpretation of structural shocks. I propose the Pandemic Priors, which augment a Bayesian VAR with time dummies for the extreme periods and shrink their coefficients under a Gaussian prior. The degree of downweighting is governed by shrinkage hyperparameters chosen by the data, nesting the exclusion of the extreme observations and their treatment as ordinary data as boundary cases, and remains compatible with conventional structural identification. An extension to a linearized DSGE allows it to see through the pandemic via its measurement equation..