If you want a simple copying downwards technique,
it's documented at
http://www.stata.com/support/faqs/data/missing.html
e.g.
gen long order = _n
bysort id (order) : replace summmig = summmig[_n=1] if mi(summmig)
Nick
[email protected]
[email protected]
> You are right on the complications which arise from the
> fact that there may
> be more than one migrant in a household (case of -id- 102 1003).
> However one interesting point is that observations in
> -summmig- are exactly
> in the same order as migrants in -idmig-. May this helps
> for question 1?
> I think the whole problem arise because I use -id- as
> common identifier when
> merging the data set containing -summmig- with my master
> data set, then as
> it appears from the sample I sent to the list, -summmig-'s
> observations
> match systematically with the first individuals in each household,
> regardless of their migration status.
> Then, I am expecting a way to re-allocate things downward.
> For instance
> considering -id- 102 1003, I want the two data points under
> -summmig- to
> match with this household's two migrants who are
> observations number 10 and
> 12 of the whole data set.
> By the way your suggestion worked fine for creating
> dummyinkind and I am
> considering re-merging using now both -id- and -idmig- as
> identifiers if
> this may help for question 1.
*
* 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/
