I don't think -swor- is quite defuct, or even defunct.
Nor is age as such necessarily a problem with old
programs.
As you say, -sample- still lacks an equivalent of
the -generate()- option in -swor-.
As for your more general problem, it may be that
the newisg -stable- option of -sort- is the key.
I've never tackled survey sampling.
Nick
[email protected]
Gabriel Demombynes
Yes, I see that sample has been updated to allow for
drawing fixed numbers of obs, not just percentages. It
seems like a better idea to adjust my routine to use
-sample- rather than keep trying to figure out why my
data doesn't behave properly with a defuct .ado file
from an old STB.
Unfortunately, -sample- is missing one nice feature of
-swor-, the ability to mark selected observations
rather than delete them. This was useful to me because
what I am doing is drawing a sample w/o replacement,
using probability weights for the sampling.
*
* 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/
