I see. The original suggestion was
replace Id = Id60 if Id60 < .
duplicates drop Id, force
It seems that you need:
replace Id = Id60 if Id60 < .
bysort Id: drop if Wave == 60 & _N == 2
Nick
[email protected]
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of [iso-8859-15] Markus K�ller
Sent: 25 February 2008 15:24
To: [email protected]
Subject: st: [iso-8859-15] Delete data that appears "twice" in a mixture of panel[iso-8859-15] and crosssectional data - part 2
Thank you for your answer, Nick, but unfortunately it doesn�t do what it should. That�s why I post my problem again - sorry for bothering you all with it again.
Maybe I didn�t explain the problem good enough in my first post:
-> I am using a mixture of paneldata and cross-sectional data which is organized in the following form:
Id Wave Age Gender ..... etc... Id60
1 60 .
2 60 .
3 60 .
4 65 .
5 65 1
6 65 .
.
.
.
etc....
To continue my analysis, I will have to delete the households from wave "60", which also appear in the "later" wave.
!!! It is important, that it has to be deleted from 60, not from 65.!!! In Nicks answer, the data gets deleted from wave 65. But since I want to use the "fresher" or "younger" data, I want to keep the data from 65 and delete the data from 60.!!!
In other words, I have to delete the data in a way, that the above table looks like:
Id Wave Age Gender ..... etc... Id60
2 60 .
3 60 .
4 65 .
5 65 1
6 65 .
.
.
.
etc....
, because household "1" also appears in 65-wave.
*
* 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/
