These data and MATLAB programs will allow you to reproduce the statistical results presented in "Money and Output: Friedman and Schwartz Revisited" by Michael T Belongia and Peter N Ireland.

Data on GDP, the GDP deflator, the federal funds rate, the Divisia and simple-sum M1, M2, and MZM monetary aggregates, the Divisia M1 user cost, and the commodity price index are contained in the text files pyr.dat, divm.dat, ssm.dat, divu.dat, and cp.dat, which can also be read directly into MATLAB. For details on the organization of these data files, see the enclosed "data guide.txt" file.

The MATLAB program ypmcors.m computes the correlations between output and money and prices and money shown in tables 1-4 of the paper. Open the file with MATLAB, select the options described in the comment section at the top of the file, the run the program from the command window.

The MATLAB program svarest6.m estimates the VARs to produce the results shown in tables 5-10 and figures 4-6. Open the file in MATLAB, select the options described in the comments section at the top of the file, and run the program from the command window.

The remaining MATLAB programs are functions called by ypmcors.m or svarest6.m. bkfilter.m is used by ypmcors.m to filter the data when they are in log levels. cllfn6.m contains the concentrated log likelihood function for the structural VAR and is repeatedly maximized by svarest6.m in order to find the global maximum. imps.m and vardecomp.m are used by svarest6.m to compute the impulse responses, variance decompositions, and their standard errors.
