Testing Big Data in a Big Crisis: Nowcasting under COVID-19
Authors:
- Luca BarbagliaLBLuca Barbaglia
- Lorenzo FrattaroloLFLorenzo Frattarolo
- Luca OnoranteLOLuca Onorante
- Luca Tiozzo PezzoliLPLuca Tiozzo Pezzoli
- Filippo M. PericoliFPFilippo M. Pericoli
- Marco RattoMRMarco Ratto
Preprint
- 30 March 2022
- preprint
- Published by Elsevier in SSRN Electronic Journal
Abstract
During the COVID-19 pandemic, economists have struggled to obtain reliable economic predictions, with standard models becoming outdated and their forecasting performance deteriorating rapidly. This paper presents two novelties that could be adopted by forecasting institutions in unconventional times. The first innovation is the construction of a large data set for macroeconomic forecasting in Europe. We collect more than a thousand time series from conventional and unconventional sources: we complement traditional macroeconomic variables with timely big data indicators, and assess their added value at nowcasting. The second novelty consists of a methodology to merge an enormous amount of non-encompassing data and a large battery of classical and more sophisticated forecasting methods in a seamlessly dynamic Bayesian framework. Specifically, we introduce an innovative "selection prior'' that is used not as a way to influence model outcomes, but as a selecting device among competing models. This allows the search for better specifications to go hand in hand with the forecasting process. By applying this methodology to the COVID-19 crisis, we show which variables are good predictors for nowcasting purposes and draw lessons for dealing with possible future crises.Keywords
This publication has 37 references indexed in Scilit:
- Predicting the Present with Google TrendsEconomic Record, 2012
- Bayesian averaging, prediction and nonnested model selectionJournal of Econometrics, 2012
- Nowcasting Business Cycles Using Toll DataJournal of Forecasting, 2011
- Online Prediction Under Model Uncertainty via Dynamic Model Averaging: Application to a Cold Rolling MillTechnometrics, 2010
- Forecast combination and the Bank of England's suite of statistical forecasting modelsEconomic Modelling, 2008
- Nowcasting: The real-time informational content of macroeconomic dataJournal of Monetary Economics, 2008
- Determinants of Long-Term Growth: A Bayesian Averaging of Classical Estimates (BACE) ApproachAmerican Economic Review, 2004
- Macroeconomic Forecasting Using Diffusion IndexesJournal of Business & Economic Statistics, 2002
- Testing the equality of prediction mean squared errorsInternational Journal of Forecasting, 1997
- Bootstrap Prediction Intervals for AutoregressionJournal of the American Statistical Association, 1990