Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: ivprobit not converge


From   Maarten buis <[email protected]>
To   [email protected]
Subject   Re: st: ivprobit not converge
Date   Wed, 18 Aug 2010 06:22:16 +0000 (GMT)

--- On Wed, 18/8/10, xueliansharon wrote:
> I want to estimate an ivprobit model, but it doesn't
> converge, no matter I use the maximization tech as dfp,
> bfgs, nr or bhhh, so what can I do to make the model
> converge? 

Stata is telling you that your data does not contain the
necessary information to estimate your model. So you need
to find out what part of your model caused your model to
be too demanding. I would start with the most minimal 
model I can come up with. I don't have to believe that that 
model makes sense, in fact I probably don't believe that
it does, but it gets the process started. Than step by
step I add variables, till my model no longer converges.
This gives me and idea what the problem is, and often 
I see right than that either the model I tried to estimate 
did not make sense or really could not be estimated with
my data. The things I am looking for than are near perfect
correlation with some of the other explanatory variables,
whether that variable is (nearly) a perfect linear 
combination of other variables (e.g. period - age = cohort),
very small categories when it is a categorical variable, 
association with the dependent variable that is so strong 
that virtually noone failed after a certain value.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


      

*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index