Dropping observations if entity does not exist in both waves

[Dickson Agbaji]

Hi,

I am new to Stata and trying to undertake a task for a stat class.

I have a dataset containing two waves of a survey, and I want to keep only observations available in both wave1 and wave2.

Please see below for an example of the data.

* Example generated by -dataex-. To install: ssc install dataex
clear
input long hhid double hhpid byte indiv float wave byte sex
10001 1000103  3 2 1
10001 1000108  8 2 0
10002 1000203  3 1 0
10002 1000208  8 2 1
10002 1000209  9 2 1
10003 1000303  3 1 0
10003 1000304  4 1 1
10003 1000305  5 1 0
10003 1000306  6 1 0
10003 1000306  6 2 0
10004 1000405  5 2 1
10005 1000503  3 2 0
10005 1000504  4 2 1
10005 1000505  5 2 1
10008 1000804  4 1 1
10008 1000805  5 1 1
10008 1000806  6 1 0
10008 1000807  7 1 1
10008 1000807  7 2 1
10008 1000808  8 2 0
10009 1000903  3 2 1
10009 1000905  5 2 1
10009 1000906  6 2 1
10010 1001004  4 1 1
10011 1001105  5 1 0
10011 1001106  6 1 1
10013 1001308  8 1 1
10016 1001603  3 1 1
end
label values sex sex
label def sex 0 "0. Male", modify
label def sex 1 "1. Female", modify

[Hemanshu Kumar]

Dickson Agbaji This is not a place to get your homework assignments solved.

[Dickson Agbaji]

Thanks for your response Hemanshu Kumar,

I have been able to solve the problem.

Here is the code in case someone else faces a similar issue: by hhpid (wave), sort: keep if _N == 2 & wave[1] == 1 & wave[_N] == 2