Re: st: suest mlogit equation not found error

[natalia malancu <natalia.malancu@gmail.com>]

Hi Fernando,

to my understanding & based on what I read in previous posts about -suest- that
specification applies only when referring regress. I tried it anyway and
I'm getting the same error.

Natalia




On Fri, Jul 20, 2012 at 6:45 PM, Fernando Rios Avila <f.rios.a@gmail.com> wrote:
>
> Hi Natalia,
> I haven't personally done a Chow test using an Mlogit model, however I
> believe the right syntax is:
>  mlogit Y x1 x2 x3 if period==0, base (0)
>  est store model1
>  mlogit Y x1 x2 x3 if period==1, base (0)
>  est store model2
>
>  suest model1 model2
>  test [ model1_mean=mode2_mean]
>
> Look at example 2 of the -suest- help file.
> Fernando
>
> On Fri, Jul 20, 2012 at 10:14 AM, natalia malancu
> <natalia.malancu@gmail.com> wrote:
> > Hi all,
> >
> > I estimate a mlogit model on two different periods of the same
> > sample. Then I want to test if there is a structural change. My
> > question is: how do I do Chow Test in stata in this case?
> >
> > At this point I tried using the -suest- command as follows:
> >
> > mlogit Y x1 x2 x3 if period==0, base (0)
> > est store model1
> > mlogit Y x1 x2 x3 if period==1, base (0)
> > est store model2
> >
> > suest model1 model2
> > test [ model1=mode2]
> >
> > but I keep receiving the following error:
> >
> > "equation model1 not found"
> >
> > I have also tried test[model1_Y=model2_Y] with similar limited success.
> >
> > Thanks for helping,
> > Natalia
> > *
> > *   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/