Conditional subsetting of records based on dates

[NewUsrStat]

Hi guys, 

suppose to have the following: 

ID               Year         

 A              2018

 A              2018

 A              2020

 A              2020

 A              2021

 B              2020

 B              2020

.......           .............

Is there a way to get the following? 

ID               Year         

 A              2018

 B              2020

.......           .............

Meaning: if ID has more than one year like "A" then take the most ancient and take it one time; if ID has more records of the same year, then take the ID and the relative year one time. 

Thank you in advance

[antonbcristina · Accepted solution]

You can achieve this with a simple DATA step.

You'll first need to sort the data by ID and YEAR.

proc sort data=have;
 	by id year;
run;

Then use a DATA step with a BY statement, keeping only the first occurence of each ID. Since the data is sorted by ID and YEAR, you'll automatically keep the earliest occurence of each ID.

data want;
	set have; 
	by id;
	if first.id;
run;