Philippe Goulet-Coulombe, Maxime Leroux, Dalibor Stevanovic, and
Stéphane Surprenant, "How is Machine Learning Useful for Macroeconomic
Forecasting?", Journal of Applied Econometrics, Vol. 37, No. 5,
2022, pp. 920-964.

The data used in this paper are publicly available from the Fred and
Alfred website, from Sydney Ludvigson's website, and from the
Dataverse of the Large Canadian Database for Macroeconomic Analysis
(LCDMA).

All files are zipped in the file glss-files.zip. This contains a mix
of binary and ASCII files.

Folder MainAnalysis contains data for the main forecasting exercise
(sections 4 to 7). These time series are in levels, so the appropriate
transformation must be applied following instructions in the paper and
McCracken and Ng(2016) for FREDMD data (2018-01.csv).

Folder SupplementaryMaterial contains the FREDMD vintages used in the
real time exercise in Appendix B.1 with forecasted series vintages
from Alfred for the different releases. It also contains data for
robustness checks using FREDQD and Canadian data.
 
Copyright: Please cite the paper properly when using any part of this
package. The usual disclaimer applies.

Dalibor Stevanovic
dstevanovic.econ [AT] gmail.com