Dropping observations without corresponding partnerID

[Juliane Dold]

!Sorry in advance, I haven't been able to extract the exact example with dataex!
Stata 18 on Windows


My sample looks roughly like that:

Year	HouseholdID	PersonalID	PartnerID
2005	10	107	108
2006	10	107	108
2007	10	107	108
2008	10	107	108
2009	10	107	108
2005	11	113	-

People living in the same household as their partner get assigned their partner's PersonalID as PartnerID. As shown in the example above, some people (107) have an assigned partner (108) but their partner isn't part of this sample (PersonalID 108 is missing). I now want to drop all the people whose partners are not part of my dataset (in this case: dropping 107).

I have tried things like:
gen partner_missing = missing(PartnerID)
drop if partner_missing
drop partner_missing

but this only drops observations like 113 which are not supposed to be dropped.

What else can I try to yield my desired result?
Thanks in advance!

[Andrew Musau]

* Example generated by -dataex-. For more info, type help dataex
clear
input int year byte householdid int(personalid partnerid)
2005 10 107 108
2006 10 107 108
2007 10 107 108
2008 10 107 108
2009 10 107 108
2005 11 113   .
end

frame put personalid, into(found)
frame found: contract *
frlink m:1 partnerid, frame(found personalid)
drop if missing(found) & !missing(partnerid)
frame drop found
drop found

Res.:

. l

     +---------------------------------------+
     | year   househ~d   person~d   partne~d |
     |---------------------------------------|
  1. | 2005         11        113          . |
     +---------------------------------------+

[Juliane Dold]

worked out perfectly, thanks so much!