Delete row with condition

[zihdonv19]

Hi!

For unique id, I want to keep the record; for non-unique id, I want to keep the records with non-missing date. Could anyone help with it? Thank you!!

What I have:

id	date
1	.
2	.
2	2000-06-10
2	2002-08-26
3	2010-02-03
4	.
4	2000-06-10
4	2005-07-26
4	2016-04-03

What I want:

id	date
1	.
2	2000-06-10
2	2002-08-26
3	2010-02-03
4	2000-06-10
4	2005-07-26
4	2016-04-03

[SASJedi · Accepted solution]

Try this:

data have;
	infile datalines truncover;
	input id date:yymmdd10.;
	format date yymmdd10.;
datalines;
1	.
2	.
2	2000-06-10
2	2002-08-26
3	2010-02-03
4	.
4	2000-06-10
4	2005-07-26
4	2016-04-03
;

/* The data must be sorted at least by ID */
proc sort data=have;
	by id date;
run;

data want;
	set have;
	by id;
	if (first.id and last.id)
		or not missing(date);
run;

My result:

Want

Obs	id	date
1	1	.
2	2	2000-06-10
3	2	2002-08-26
4	3	2010-02-03
5	4	2000-06-10
6	4	2005-07-26
7	4	2016-04-03