Ed Herbst and Frank Schorfheide, "Sequential Monte Carlo Sampling for 
DSGE Models", Journal of Applied Econometrics, Vol. 29, No. 7, 2014, pp. 
1073-1098.

There are two datasets. Both are stored in ASCII files in DOS format,
which are zipped in the file hs-data.zip. Unix/Linux users should use
"unzip -a".


sw_actual.txt -- This contains the data used in the in the estimation
  of the Smets-Wouters model.  There are 7 series from 1966Q1 -- 2006Q4:
 
  1. Per capita real output growth.
  2. Per capita real consumption growth.
  3. Per capita real investment growth.
  4. Per capita real wage growth.
  5. Log Per capita hours level (computed as an index.)
  6. Quarterly inflation computed via the GDP deflator.
  7. Quarterly nominal federal funds rate. 

Additional details can be found in:

  Frank Smets and Raf Wouters (2007): "Shocks and Frictions in US
  Business Cycles," American Economic Review, 97, 586-608.


sgu_actual.txt -- This contains to data used in the estimation of the
  Schmitt-Grohe-Uribe model.  There are 7 series from 1955Q1 -- 2006Q4:

   1. Real GDP growth.
   2. Real consumption growth.
   3. Real investment growth.
   4. Real government spending growth.
   5. Hours.
   6. TFP growth.
   7. The relative price of investment. 

Additional details can be found in:
  
  Stephanie Schmitt-Grohe and Martin Uribe (2012): "What's News in
  Business Cycles," Econometrica, 80, 2733-2764.


Fortran and Matlab code for this paper is available at:
  http://edherbst.net