Re: st: IV-oprobit using the cmp command

[Benjamin Volland <volland@econ.mpg.de>]

Sorry, no.
A quick glance over older messages to the list, revealed e.g. this:
http://www.stata.com/statalist/archive/2010-04/msg01375.html
You can check whether you've installed the most recent version of -cmp-.
Otherwise it may also be interesting to the more advanced users whether 
you can replicate the error with a data set available from within stata.
If I run (using stata 12)
. sysuse auto.dta
. cmp setup
. cmp (mpg=trunk) (trunk=rep78), ind(5 5) tech(dfp) nolr
everything works fine.

Best, Ben

On 05/04/2012 17:24, Peter Gambrill wrote:
> Thanks Ben
>
> However when I follow and augment for my own case the example given in
> the document for IV-oprobit i.e. when I write this (where varlist is
> the rest of my variables/instruments)
>
> cmp (identity=varlist) (hlghq1= identity varlist), ind($cmp_oprobit
> $cmp_oprobit) tech(dfp) nolr
>
> I get an error
>
> "Fitting full model.
>
>        cmp_lnL_censored():  3301  subscript invalid
>                 cmp_lnL():     -  function returned error
>                   <istmt>:     -  function returned error
> Mata run-time error
> r(3301);"
>
> Any ideas?
>
> Many Thanks
>
>
> On Thu, Apr 5, 2012 at 3:53 PM, Benjamin Volland<volland@econ.mpg.de>  wrote:
> > Hello Peter,
> >
> > David Roodman, the author of -cmp- (user-written), provides an extensive
> > discussion of this command (including statistics and instructions),
> > available as working paper on the website of the Center for Global
> > Development:
> > http://www.cgdev.org/files/1421516_file_Roodman_cmp_FINAL.pdf
> >
> > Best, Ben
> >
> >
> >
> > On 05/04/2012 16:35, Peter Gambrill wrote:
> > > I am trying to using Instrumental variables using a probit
> > > specification. The command I have found is the "cmp" command, I just
> > > wondered if anyone had experience of using this and could help me with
> > > this estimation.
> > >
> > > My endogenous variable is a measure of gender identity. This is a
> > > discrete variable that has a scale of 7 to 33.
> > >
> > > Gender Identity = constant + instruments + error
> > >
> > > The second stage dependant variable is a measure of subjective well
> > > being on a scale of 1 to 36
> > >
> > > Subjective well-being = constant + controls + gender identity + error
> > >
> > > Can anyone give some advice on how I might achieve this estimation
> > > using stata?
> > >
> > > Many Thanks
> > >
> > > Peter
> > > *
> > > *   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/
> >
> > *
> > *   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/
>
> *
> *   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/

*
*   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/