--- On Wed, 10/2/10, Sun, Yan (IFPRI) wrote:
> I would like to do linear hypothesis for Probit. I know how
> to do it for OLS. For example, for the equation y=a+b*x1+c*x2,
> I would use STATA command: "test (b=0) (c=0)", or "test b=c".
>
> But now y is a bivariate dummy and I would use probit. How
> could I use STATA command to do a joint F test? I find there
> is "test" command after probit. Is this "test" command after
> probit a correct command for a linear hypothese test for
> probit?
-test- is fine after -probit-, it will give you the Wald test.
An alternative is the likelihood ratio test, which has some
nice characteristics, but cannot be used when you specified
the -robust-, -cluster()- options or used pweights. Below is
an example showing how to do both.
*------------ begin example ---------------
sysuse auto, clear
probit foreign rep78 price mpg
test (rep78 = 0) (price = 0)
est store a
probit foreign mpg if !missing(rep78,price)
est store b
lrtest a b
*----------- end example -------------------
( For more on how to use examples I sent to statalist see:
http://www.maartenbuis.nl/stata/exampleFAQ.html )
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/
