Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Non-Converging Likelihood in a Structural VAR


From   "Downey, Patrick" <PDowney@urban.org>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Non-Converging Likelihood in a Structural VAR
Date   Wed, 10 Mar 2010 10:47:22 -0500

Hello,

I'm trying to estimate a structural VAR, and I let the program run for
about an hour the other night, but it didn't converge to a maximum
likelihood.

My code is:
matrix input aconst1 = (1,.,. \ .,1,. \ .,.,1)
matrix input bconst1 = (1,.,. \ .,1,. \ .,.,1)
matrix input aconst2 = (.,12,13 \ 12,.,23 \ 13,23,.)
matrix input bconst2 = (.,12,13 \ 12,.,23 \ 13,23,.)
svar cgtargetall cg1000all cg2000all, aeq(aconst1) acns(aconst2)
beq(bconst1) bcns(bconst2) lags(1 2) exog(famunitocc familyunitstable
noresid)

When I had the wrong number of constraints, Stata gave me an error saying
so. It's no longer giving the error, so presumably that's not the problem.
Also, my dataset is fairly small (T = 117), so I don't think that's the
problem either. I've tested each series independently (previously) and they
are all stationary. Finally, I haven't tested for cointegration, but I
don't imagine that I have that problem, and if I do, I'm not sure that
would cause non-convergence in the likelihood. Are there any other possible
causes?

Any guidance is greatly appreciated.


P. Mitchell Downey | Research Associate II
Justice Policy Center | Urban Institute
2100 M. Street N.W. | Washington, DC 20037
T: (202) 261-5329 | F: (202) 659-8985
PDowney@urban.org | www.urban.org


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index