Agreed, with one wrinkle:
gen long obs = _n
bysort county precinct (obs) : keep if _n == 1
keeps the original observation numbers if they
are important to you.
Nick
[email protected]
Michael Blasnik
> Given the large dataset, I would just drop all of the cases
> after the first
> for each criteria grouping, the you can list whatever you
> like quickly and
> easily:
>
> bysort county precinct: keep if _n==1
>
> Then you could:
>
> list if county==xxx & precinct==y
>
> It will be much faster working with the reduced dataset.
> Just remember not
> to save,replace.
paul gronke
> >I am dealing with an extremely large (2 gig) dataset and wish to list
> > just the first case that meets a set of criteria. There
> are about 200
> > separate sets of criteria (county and precinct values).
> >
> > Thus, I'd like a command that does something like:
> >
> > list name address1 address2 county precinct city state
> > if [first one of] county=VALUE & precinct=VALUE
> >
> > repeated about 200 times.
> >
> > I could do this with
> > sample 1 if CRITERIA
> >
> > but that seems needless complex, especially since all I want is the
> > first case that meets 200 combinations of county and precinct.
*
* 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/
