Volume 20, Issue 4 October 2024

Real-Time Forecasting with a (Standard) Mixed-Frequency VAR During a Pandemic

Abstract

We resuscitated the mixed-frequency vector autoregression (MF-VAR) developed in Schorfheide and Song (2015) to generate macroeconomic forecasts for the United States during the COVID-19 pandemic in real time. The model combines 11 time series observed at two frequencies: quarterly and monthly. We deliberately did not modify the model specification in view of the COVID-19 outbreak, except for the exclusion of crisis observations from the estimation sample. We compare the MF-VAR forecasts to the median forecast from the Survey of Professional Forecasters (SPF). While the MF-VAR performed poorly during 2020:Q2, subsequent forecasts were at par with the SPF forecasts. We show that excluding a few months of extreme observations is a promising way of handling VAR estimation going forward, as an alternative of a sophisticated modeling of outliers.

Authors

  • Frank Schorfheide
  • Dongho Song

JEL codes

  • C11
  • C32
  • C53

Other papers in this issue

Toni Ahnert and Katrin Assenmacher and Peter Hoffmann and Agnese Leonello and Cyril Monnet and Davide Porcellacchia

Shalva Mkhatrishvili and Giorgi Tsutskiridze and Lasha Arevadze

Tobias Adrian and Vitor Gaspar and Francis Vitek

Laura Acevedo and Marc Hofstetter

Juan M Londono and Stijn Claessens and Ricardo Correa