duplicates drop id, force
will leave you with panels with one observation
each. At least that is my understanding of
-duplicates-.
I don't find this well specified. If you
want a random subset of observations, you need
to do nothing special. If you want a random
subset of panels, you need to say more about
what kind of random subset.
Nick
[email protected]
chen jie
> I have a problem on how to select observations from a
> unbalanced panel data?
>
> I mean, the original data is too big and I want only a
> smaller section. If it is a balanced panel data, I can
> for example choose 50% random part. But now it is a
> unbalance panel data, I can't do that otherwise the
> new data will suffer from "length bias". I think
> first I need to randomly choose 50% of individuals
> (duplicates drop id, force==>sample 50), but how I can
> proceed next by using this unique id key index to
> produce new panel data?
*
* 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/
