In the excerpt of data you present, each T variable contains only a single
entry for A, the others being all missing. If that is true overall (and
you should verify this with appropriate assert statements before
proceding) you can get what you are looking for with:
-collapse (sum) T*, by(var)-
Good luck.
Clyde Schechter
Albert Einstein College of Medicine
Bronx, New York, USA
> - -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Mark Pachucki
> Sent: Dienstag, 8. Dezember 2009 23:35
> To: [email protected]
> Subject: st: combining values distributed over multiple observations
>
> Hello - this seems like an exceedingly simple question, but couldn't find
> a
> reference to relevant syntax in the help. What I'm essentially trying to
> do
> is go from the below format, where a given variable "A" takes different
> values at discrete observations arrayed over T1-T5:
>
> var T1 T2 T3 T4 T5
>
> A .129 . . . .
> A . .647 . . .
> A . . .991 . .
> A . . . .582 .
> A . . . . .043
>
>
> to the following format, wherein all values of A at each time are
> compressed
> into one observation :
>
> var T1 T2 T3 T4 T5
>
> A .129 .647 .991 .582 .043
>
> It seems to be at the intersection of ideas behind reshape / stack (and
> even
> stjoin - except it's not survival-time data). Yet doesn't fall neatly
> within
> any (or does it?)
>
> Thanks in advance,
> Mark
*
* 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/
