Tae-Hwy Lee, Shahnaz Parsaeian, and Aman Ullah, "Optimal Forecast under 
Structural Breaks", Journal of Applied Econometrics, Vol. 37, No. 5,
2022, pp. 965-987.

The data used in the paper are available with the GVAR toolbox,
Mohaddes and Raissi (2018).

This dataset is taken from
https://sites.google.com/site/gvarmodelling/data. We use the GVAR Data
1979Q2-2016Q4 (2016 Vintage). The data were updated by Mohaddes and
Raissi (2018). We focus on the 9 countries: Australia, Canada, France,
Germany, Italy, Japan, Spain, UK, and USA. 

The dataset for each country is saved as an ASCII (text) file in DOS
format. All files are zipped in lpu-data.zip. Unix/Linux users should
use "unzip -a".

The sample spans the period 1979Q2 to 2016Q4 (151 observations).  

The variables are the lag of the real GDP (yt-1), real equity prices
(eq), real short term interest rate (r), the difference between the
real long term interest rate and the real short term interest rate
(lr-r). The corresponding country-specific foreign variables are
denoted by yt-1*, eq*, r*, and (lr-r)*, respectively. The foreign
variables are constructed using rolling three-year moving averages of
the annual trade weights, which are computed as shares of exports and
imports for each country. The data are for the one-step-ahead forecast.

The detailed description of data is available in GVAR Database
(1979Q1-2016Q4).pdf and Readme.pdf provided by Mohaddes and Raissi
(2018), which can be downloaded from
https://sites.google.com/site/gvarmodelling/data.

****

Mohaddes, K., and Raissi M. (2018). Global VAR (GVAR) database,
1979Q2-2016Q4. https://doi.org/10.17863/CAM.27337.