Recovering parameter values after moments (or IRF) calibration

Dear all.

While discovering the uses of the GSA toolbox I have been struck by a question.

Following the example of Ratto, Hohberger and Burgert (2016) (, are the parameter value combinations that appear in Figure 7 stored in some place? If not, is there any way I can retrieve them?

I have already looked at the documentation and at the diverse example codes that are scattered over the Internet, but I have not been able to find an answer as of yet.

Thank you in advance.

I think I may have found the answer to my question.

For example, just focusing on prior restrictions, a [modname]_prior.mat file is created in the “gsa” folder. In this matfile, there are three objects of interest: lpmat, lpmat0 and irestriction. The first object appears to contain the values of the parameters corresponding to each iteration, the second object seems to contain the values of the standard deviations of the shocks also in each iteration, while the third object shows which are the iterations that comply with the imposed moment and/or IRF restrictions.

I would be very grateful if somebody could, please, validate my statement.

more detailed info is in the file
[your model]_prior_restrictions.mat

there, the entries of each column indx_irf(:,j) ==0 and indx_moment(:,j)==0 provide the rows of lpmat that match the restriction for each row j of the list of irf/moment restrictions define.

The list of restrictions is also stored in the file [your model]_prior_restrictions.mat so that each row j of


report info on each restriction that map on each j column of indx_irf(:,j) and indx_moment(:,j)

indeed you are right. the irestriction indices tell you params which fulfill ALL restrictions simultaneously.
the more detailed info I mentioned earlier, allow you to inspect info for each individual restriction

Thank you very much for your answer.

Is there any documentation in which the contents of the other matfiles (and also the other features of the toolbox) are explained?

unfortunately not (yet …)

Thanks once again. I will be looking forward to learning more about the GSA toolbox.