Assuming that X1 contains only 0 and 1 and has no missings, you can go on as
follows:
. by UIDhhstr (X1), sort: gen x2 = x1[_N]
If X1 also has other values than 0 and 1 or has missings, the following should
work:
. by UIDhhstr, sort: gen x2 = sum(X1==1)
. by UIDhhstr: replace x2 = x2[_N]>=1
My general advice here is to learn recoding with -by varlist:- and explicit
subscripts. A possible starting point would be
SJ-2-1 pr0004 . . . . . . . . . . Speaking Stata: How to move step by: step
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . N. J. Cox
Q1/02 SJ 2(1):86-102 (no commands)
explains the use of the by varlist : construct to tackle
a variety of problems with group structure, ranging from
simple calculations for each of several groups to more
advanced manipulations that use the built-in _n and _N
(or take a look at pg. 80-86 in "Data Analysis Using Stata" by Frauke Kreuter
and myself, if you allow me that advertisement.)
many regards
Uli
On Thursday 06 October 2005 19:58, F V wrote:
> I have a very large data-set (i.e. around 60,000
> households, and around 320,000 individuals). say that
> i have three variables: a hh identifier, a person
> identifier, and an X1 variable of interest that
> measures whether a person is X1 or not.
>
>
> UIDhhstr X1 UIDpersonstr
> 01101021 1 0110102101
> 01101021 0 0110102102
> 01101021 0 0110102103
> 01101031 1 0110103101
> 01101031 0 0110103102
> 01101031 0 0110103103
> 01101031 0 0110103104
> 01101031 0 0110103105
> 01101031 0 0110103106
>
>
> say I wanted to tell Stata: gen a variable X2 such
> that it takes the value of 1 if for any member of a
> household X1==1. how can i do this in stata? after
> generating X2, then i could collapse the data down to
> the household level.
--
[email protected]
+49 (0)30 36991040
*
* 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/
