topic Subset data based on the value included in the repeated observations in SAS Programming

Subset data based on the value included in the repeated observations

Cruise — Mon, 09 Apr 2018 19:07:52 GMT

I have patients in my data with repeated observations. "uid"=unique identifier of patients. dos=date of survey. I'd like to subset data where patients have future date of survey (2019). Below approach outputs the rows only where dos=2019 but not along with previous years of information which I need to have to investigate the source of error.

Your help is appreciated to come up with data "want" from the code block below. Using SAS 9.4.

data have;
input uid dos;
cards;
1 2015
1 2016
1 2019
2 2017 
2 2018
3 2019
4 2015
4 2016
4 2017
5 2015
;

data want;
input uid dos;
cards;
1 2015
1 2016
1 2019
3 2019
;

proc sort data=have; 
by uid;
run; 
data wrong; set have;
by uid;
if dos in (2019) then output;
run;

Re: Subset data based on the value included in the repeated observations

Reeza — Mon, 09 Apr 2018 19:15:54 GMT

proc sql;

create table want as
select * from have 
where uid in 
    (select uid from have where dos=2019);

quit;

Re: Subset data based on the value included in the repeated observations

Cruise — Mon, 09 Apr 2018 19:24:10 GMT

worked out! but why? trying to visualize how SAS is calling data here

Re: Subset data based on the value included in the repeated observations

Reeza — Mon, 09 Apr 2018 19:26:08 GMT

proc sql;

create table want as
select * from have 
where uid in /*selects only IDS from this list*/ 
    (select distinct uid from have where dos=2019); *<- this creates a list of IDs where it is 2019, should add the DISTNCT word here;

quit;