The files in this directory were used in my evaluation
of the SVAR procedure of 

   Jordi Gali and Pau Rabanal 
   ``Technology Shocks and Aggregate Fluctuations:
     How Well Does the RBC Model Fit Postwar US Data?''
     NBER Macroeconomic Annual 2004

The main codes generating the results are monte.m (for
the benchmark model) and monte_ur.m (for the extension 
with random-walk technology).  These codes take simulated
data from the RBC models and apply GR's SVAR procedure. 

Ellen McGrattan
June 2004


FILE             BRIEF DESCRIPTION
----             -----------------
appendix1.*      Appendix describing computations preformed
bca.m            Compute statistics used in ``Business Cycle Accounting''
draw*.m          Draw sequences of shock innovations and generate time series
fake_gr.m        Apply SVAR procedure to data in fakedata.dat
fakedat.dat      Model data given to SVAR researcher
figure2.dat      Hours figure computed with monte.m
figure3.dat      Technologies from bca.m (log(zt)), ../replicte/drawz.m (lz)
mle*.m           Subroutines used for MLE estimation
monte.m          Main code for benchmark model
monte_ur.m       Main code for model with random-walk technology
nipa*.csv        NIPA data files
prescott.*       Population and hours data file and description
res_*.m          Subroutines with first-order conditions for the models
runmle*.m        Drivers for MLE estimation
setupdata.m      File used to set up input files for MLE
uszvarq*.dat     Output of setupdata.m