It doesnt work
global malesuic (divper_lag income_lag gini_xpg alcohtot gunhomper
unempl p1019 p65 pctcatholic pcprotestant pctmormon pctbabtispt
i1-i51 )
global femalesuic (divper_lag income_lag gini_xpg alcohtot gunhomper
unempl p1019 p65 pctcatholic pcprotestant pctmormon pctbabtispt
i1-i51 )
sureg $malesuic $femalesuic
I got the message singular matrix, i am trying to estimate a fixed effects model
test [malesuic] gunhomper = [femalesuic] gunhomper
On Jan 14, 2008 4:57 PM, Kit Baum <[email protected]> wrote:
> Stata is unable to see that your test of x==x makes no sense... In
> the suest context, this is a multiple-equation system, so you must
> compare gunhomper in the malesuic equation with that in the
> femalesuic equation. Check -help test- for how to specify constraints
> in a multiple-equation system, i.e. in the -sureg- context.
>
>
> Kit Baum, Boston College Economics and DIW Berlin
> http://ideas.repec.org/e/pba1.html
> An Introduction to Modern Econometrics Using Stata:
> http://www.stata-press.com/books/imeus.html
>
>
> On Jan 14, 2008, at 10:52 AM, antonio rodriguez wrote:
>
> > I got this result, there must be something wrong, right?
> >
> > ***Fixed effects with the dependent variable (male suicide rate)***
> >
> > regress malesuic divper_lag income_lag gini_xpg alcohtot gunhomper
> > unempl p1019 p65 pctcatholic pcprotestant pctmormon pctbabtispt
> > i1-i51 [aw=pop]
> > estimates store male
> > regress femalesuic divper_lag income_lag gini_xpg alcohtot gunhomper
> > unempl p1019 p65 pctcatholic pcprotestant pctmormon pctbabtispt
> > i1-i51 [aw=pop]
> > estimates store female
> >
> > suest male female
> > test gunhomper = gunhomper
> > test gunhomper = gunhomper
> >
> > ( 1) = 0
> > Constraint 1 dropped
> >
> > chi2( 0) = .
> > Prob > chi2 = .
> >
> > .
> > end of do-file
> >
> >
> > On Jan 14, 2008 2:13 PM, Kit Baum <[email protected]> wrote:
> >> Antonio
> >>
> >> If your message is not going through to Statalist it is because you
> >> are not subscribed to Statalist from this address or you are sending
> >> it in some form other than straight ASCII.
> >>
> >> The areg command is not supported by suest. It does not seem that the
> >> logical equivalent, xtreg,fe, is either. If the number of values of
> >> state is not too large, you could perform the within transformation
> >> "by hand" and use regress on each equation, and suest thereafter.
> >> Alternatively you could use the demeaned data in sureg, which would
> >> support the tests you want.
> >>
> >> Best wishes
> >> Kit
> >>
> >> Kit Baum, Boston College Economics and DIW Berlin
> >> http://ideas.repec.org/e/pba1.html
> >> An Introduction to Modern Econometrics Using Stata:
> >> http://www.stata-press.com/books/imeus.html
> >>
> >>
> >> On Jan 14, 2008, at 7:57 AM, antonio rodriguez wrote:
> >>
> >>> Dear Kit
> >>>
> >>> I am just trying to post this message on the STATA list but the
> >>> email
> >>> gets back to me. I dont know if you could help me with my
> >>> question or
> >>> forward it to stata list
> >>>
> >>> I just want to test wheter there is a statistical difference between
> >>> the same variables for different regressions with different
> >>> dependent
> >>> variables
> >>>
> >>> areg femalesuic divper_lag income_lag gini_xpg alcohtot gunaccid
> >>> unempl p1019 p65 pctcatholic pcprotestant pctmormon pctbabtispt
> >>> [aw=pop], absorb(state)
> >>> estimates store est_1
> >>>
> >>> areg femalesuic_re divper_lag income_lag gini_xpg alcohtot
> >>> gunaccid_re unempl p1019 p65 pctcatholic pcprotestant pctmormon
> >>> pctbabtispt [aw=pop], absorb(state)
> >>> estimates store est_2
> >>>
> >>> suest est_*
> >>>
> >>> test gunaccid = gunaccid_re
> >>>
> >>> --
> >>> Antonio Rodr�guez Andr�s
> >>> Profesor Visitante
> >>> Departamento de Economia y Finanzas
> >>> Universidad de Castilla La Mancha
> >>> Albacete, Espa�a
> >>
> >>
> >
> >
> >
> > --
> > Antonio Rodr�guez Andr�s
> > Profesor Visitante
> > Departamento de Economia y Finanzas
> > Universidad de Castilla La Mancha
> > Albacete, Espa�a
>
>
> *
> * 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/
>
--
Antonio Rodr�guez Andr�s
Profesor Visitante
Departamento de Economia y Finanzas
Universidad de Castilla La Mancha
Albacete, Espa�a
*
* 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/
