Christian Kascha and Carsten Trenkler, "Simple Identification and
Specification of Cointegrated VARMA Models", Journal of Applied
Econometrics, Vol. 30, No. 3, 2015, pp. 675-702.

All files are stored in the zip file kt-files.zip, which includes a
main folder and several subfolders. All text files are ASCII files
in DOS format, but there is also one binary (.xls) file. Unix/Linux
users should therefore exercise caution.


DATA:

The data used in the paper are stored in "/data/USdata.csv", which 
contains the data in ASCII format.

There are 5 data series with 516 monthly observations in time ranging
from 1970:1 to 2012:12. 
   
The series are:

    - TB 3M: 3-Month Treasury Bill: Secondary Market Rate
    - TB 6M: 6-Month Treasury Bill: Secondary Market Rate
    - GS 1 : 1-Year Treasury Constant Maturity Rate 
    - GS 5 : 5-Year Treasury Constant Maturity Rate
    - GS 10: 10-Year Treasury Constant Maturity Rate
    
The data originate from the FRED database of the Federal Reserve Bank
of St. Louis; http://research.stlouisfed.org/fred2/

In addition, there is the (binary) file "/data/USdata.xls" which is 
used in the MATLAB programs. Sheet 3 contains the data used in the
paper. 


PROGRAMS: 

All the MATLAB files to generate the main results of the paper are in 
the base folder. These are
 
make_all_tables.m:         Generate the tables.
    
make_com2RW_graph.m:       Generate graphs showing cumulative
                           sums of squared prediction errors. 
    
make_frcst_comp.m:         Compute forecasts for empirical application.
    
make_yield_graph.m:        Generate graph of all yields.
     
MCcomp_FMA_ECH.m:          Produce Monte Carlo study for comparing
                           FMA and Echelon identification.
    
MCcomp_hT_nu_choices.m:    Produce Monte Carlo study for comparing
                           parameters in order estimation. 
    
All programs should run provided the necessary input is computed
first and appropriate directories are made to save to.

The folder mfiles contains a large number of Matlab (.m) files that may 
be called by the programs in the base folder.