Re: st: Re: Generating subsamples according to a binary choice

[Michael Ingre <Michael.Ingre@ipm.ki.se>]

You have two options with the if statement in almost all stata 
commands. Note the comma -,- sign:

command xvar yvars , options , if statement
command xvar yvars if statement , options

Michael



> Thank you.
> It surely works in case of a simple regression model.
> Do you have a good idea to do the same in case of a 2step regression 
> model such as zip(zero inflated poisson) or zinb(zero inflated 
> negative binominal) where I obtain "if invalid r(198);" from Stata7?
>
>
>  --- statalist@hsphsun2.harvard.edu ---
>
>
> > --- fukugawa-nobuya@rieti.go.jp wrote:
> > > Dear Statalisters,
> > > I want to classify the sample into two parts accoding to a
> > > dichotomy variable and apply them a same regression model.
> > > For instance, I want to divide the sample into the subsample A
> > > where a dummy takes one and the subsample B where a dummy takes
> > > zero. I do this by generating two new worksheets for subsamples
> > > A&B in MS-Excel, and then run a regression twice in Stata7.
> > > How do I do these procedures using only Stata instead of generating
> > > new worksheets in Excel?
> > > Thank you in advance.
> > > Nobuya Fukugawa
> > Is this what you need?
> >
> > . sysuse auto
> > . regress mpg weight if foreign==1
> > . regress mpg weight if foreign==0
> >
> > Friedrich Huebler
>
>
> *
> *   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/
*
*   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/