I am not following this closely, but it may be that
bysort y (x) : replace x = x[_N]
is what you want.
gsort y - x
by y : replace x = x[1]
is a long-winded equivalent.
Nick
[email protected]
Hendri Adriaens
> > This should do it:
> >
> > bys y (x): replace x=x[1]
>
> Unfortunately not. Because the sorting brings the empty opnes
> to the top,
> all the matching x's are replaced by the empty (missing)
> string, which makes
> me loose more data. I tried doing it with a gsort before the
> by command:
>
> gsort y -x
>
> But then
>
> by y (x): replace x=x[1]
>
> Complains r(5): not sorted. Is there another way?
*
* 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/
