Repeated measure with duplicate observations

[desireatem]

I have this repeated measure with varying duplicate observation. I wish to make sure for each id, there is just one timepoint. Below is the data and corrected version (bold). I just one observation for each timepoint by sid

SID	NPiL	SizeL	timepoint
200001	3.8	.	1
200001	3.8	.	1
200001	3.8	.	1
200001	3.8	.	1
200001	3.8	.	1
200001	3.8	4.99	1
200001	3.8	4.89	2
200001	3.8	4.89	2
200001	3.8	4.89	2
200001	3.8	4.89	2
200001	3.8	4.89	2
200001	3.8	4.89	2
200001	4.3	4.22	3
200001	4.3	4.22	3
200001	4.3	4.22	3
200001	4.3	4.22	3
200001	4.3	4.22	3
200001	4.3	4.22	3
200001	4.2	4.25	4
200001	4.2	4.25	4
200001	4.2	4.25	4
200001	4.2	4.25	4
200001	4.2	4.25	4
200001	4.2	4.25	4
200001	4.1	5.71	5
200001	4.1	5.71	5
200001	4.1	5.71	5
200001	4.1	5.71	5
200001	4.1	5.71	5
200001	4.1	5.71	5
200001	3.9	4.96	6
200001	3.9	4.96	6
200001	3.9	4.96	6
200001	3.9	4.96	6
200001	3.9	4.96	6
200001	3.9	4.96	6
200001	4.1	5.17	7
200001	4.1	5.17	7
200001	4.1	5.17	7
200001	4.1	5.17	7
200001	4.1	5.17	7
200001	4.1	5.17	7
200001	3.5	4.71	8
200001	3.5	4.71	8
200001	3.5	4.71	8
200001	3.5	4.71	8
200001	3.5	4.71	8
200001	3.5	4.71	8
200001	3.1	5.36	9
200001	3.1	5.36	9
200001	3.1	5.36	9
200001	3.1	5.36	9
200001	3.1	5.36	9
200001	3.1	5.36	9
200001	2.4	5.81	10
200001	2.4	5.81	10
200001	2.4	5.81	10
200001	2.4	5.81	10
200001	2.4	5.81	10
200001	2.4	5.81	10
200001	3.1	5.78	11
200001	3.1	5.78	11
200001	3.1	5.78	11
200001	3.1	5.78	11
200001	3.1	5.78	11
200001	3.1	5.78	11
200001	2.9	5.5	12
200001	2.9	5.5	12
200001	2.9	5.5	12
200001	2.9	5.5	12
200001	2.9	5.5	12
200001	2.9	5.5	12
200001	2.4	6.04	13
200001	2.4	6.04	13
200001	2.4	6.04	13
200001	2.4	6.04	13
200001	2.4	6.04	13
200001	2.4	6.04	13
200001	2.6	6.24	14
200001	2.6	6.24	14
200001	2.6	6.24	14
200001	2.6	6.24	14
200001	2.6	6.24	14
200001	2.6	6.24	14
200001	3	5.42	15
200001	3	5.42	15
200001	3	5.42	15
200001	3	5.42	15
200001	3	5.42	15
200001	3	5.42	15
200001	4.2	3.58	16
200001	4.2	3.58	16
200001	4.2	3.58	16
200001	4.2	3.58	16
200001	4.2	3.58	16
200001	4.2	3.58	16
200001	2.8	6.08	17
200001	2.8	6.08	17
200001	2.8	6.08	17
200001	2.8	6.08	17
200001	2.8	6.08	17
corrected data
SID	NPiL	SizeL	timepoint
200001	3.8	.	1
200001	3.8	4.89	2
200001	4.3	4.22	3
200001	4.2	4.25	4
200001	4.1	5.71	5
200001	3.9	4.96	6
200001	4.1	5.17	7
200001	3.5	4.71	8
200001	3.1	5.36	9
200001	2.4	5.81	10
200001	3.1	5.78	11
200001	2.9	5.5	12
200001	2.4	6.04	13
200001	2.6	6.24	14
200001	3	5.42	15
200001	4.2	3.58	16
200001	2.8	6.08	17

[TomKari · Accepted solution]

Assuming your observations are sorted by SID and timepoint, and that you always want the first observation of the group, this should do it.

Tom

data want;
	set have;
	by SID timepoint;

	if first.timepoint;
run;

[SASUser_22]

proc sort data = dataset1 nodupkey;

by timepoint;

run;