Greetings statalistserv and Brian Poi,
Rest assured listservers that there is not a bug in -ivtobit-. In
setting up a file to send to Brian, I realized that it was in fact my
user error: by increasing my matsize from 600 to 800, I no longer get
the "invalid syntax" error. Extra special thanks to Brian for
requesting a copy of my data!
Adrienne
On Wed, 30 Nov 2005, Adrienne L wrote:
> Greetings statalistserv,
>
> I am using ivtobit. When I include more than 35 exogenous variables,
> I get the following error: "invalid syntax." I've tried different
> combinations of the 35 variables and have made sure that of there is a
> value in each of my variables for each observation. I am running
> Stata/SE 9.0. Originally I thought that it was incompatibility with
> the * wildcard (such a question was posted previously on this
> listserv), but if I limit my wildcard to < 35 variables, there is no
> error.
> Specifically I am running the following:
> ivtobit y _Ix2* _Ix3* _Ix4* _Ix5* _Ix6* _Ix7* (x1=x8)
> [pweight=weight], ll ul cluster(village)
>
> Neither ivreg or ivprobit (when y is converted to binary) have the
> "invalid syntax" problem
>
This could be a bug in the syntax parsing of -ivtobit-.
If Adrienne could send me (privately) a dataset and short do-file to
replicate the problem, I would be happy to look into it.
-- Brian Poi
-- [email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
