panelvar

Experimental tools on pVAR models for timely estimates.

The material provided herein will enable you to reproduce the experiments presented in Eurostat statistical working paper on New methods for timely estimates (cite this source code or the reference's doi: 10.2785/600130). Further details are also available in the other associated working papers (see Kapetanios et al.'s publications below).

Description

The source code is provided as is in the model/ folder so as to explore out-of-sample forecasting performance of mixed-frequency panel vector autoregression (pVAR) models for four key macroeconomic variables, with the goal of providing evidence on the usefulness and reliability of these models for use by official statistical agencies.

Data from four European economies, as used in the paper, are made available under the data/input folder. Output data will be stored into an data/output/ folder that needs to be created beforehand.

Additionally, the source code in tables/ enable to reproduce some figures of the working paper that are stored in the data/tables folder

Quick launch

• Run a pVar model from within a Jupyter notebook in binder (current build with commit d89d9bd). We provide the interactive environments with already installed packages to run the experiments, for instance:

• Run RStudio in binder, and simply source any of the pVar models from the platform (i.e., select the model file and run it):

Usage

To run the experiments (e.g., in your local), you will need to prior install the following package dependencies: forecast, imputeTS, panelvar, lubridate, vars, moments and zoo.

Once the packages installed, you can run the bash scripts in the bin/ folder.

About

	authors	Papailias F., Kapetanios G., Marcellino M., and Mazzi G.L.
	version	1.0
	status	2020 – closed
	license	EUPL (cite the source code or the reference above!)

References

• Kapetanios G., Marcellino M., Papailias F. and Mazzi G.L. (2020): New methods for timely estimates, Eurostat Statistical Working Paper KS-TC-20-005-EN, doi: 10.2785/600130.

• Mazzi G.L. and Mitchell J. (2020): New methods for timely estimates: nowcasting euro area GDP growth using quantile regression, Eurostat Statistical Working Paper S-TC-20-004-EN, doi: 10.2785/26603.

• Sigmund M. and Ferstl R. (2019): Panel Vector Autoregression in R with the package Panelvar, Quarterly Review of Economics and Finance, doi: 10.2139/ssrn.2896087.

• Mazzi G.L. and Ladiray D., eds. (2017): Handbook on Rapid Estimates, Publications Office of the European Union, doi:10.2785/4887400.

• Kapetanios G., Marcellino M. and Papailias F. (2017): Guidance and recommendations on the use of Big data for macroeconomic nowcasting, in Handbook on Rapid Estimates, Chapter 17, Publications Office of the European Union, doi:10.2785/4887400.

• Dees S. and Gunter J. (2014): Analysing and forecasting price dynamics across Euro area countries and sectors - A panel var approach, European Central Bank Working Paper QB-AR-14-098-EN, no. 1724.

• Croissant Y. and Millo G. (2008): Panel data econometrics in R: The plm package, Journal of Statistical Software, 27(2):1-43, doi: 10.18637/jss.v027.i02.