Calculating Min and Max

bignate1030 — Wed, 19 Jun 2019 15:28:33 GMT

I am trying to calculate Min and Max in multiple capacities with a data set.

Here is an example of the data I am working with:

person_id event id var2 var3 var4 var5 var6

1 1 2 3 1 3 5

1 2 6 1 2 4 1

2 3 3 1 4 2 5

So there are people who have multiple rows within the data set but each observation will have a unique event id.

For var2 - var6 I am trying to calculate an overall max and min for each person_id and an individualized max and min for each event_id.

Here is my desired output:

person_id event id var2 var3 var4 var5 var6 maxPERSON minPERSON maxEVENT minEVENT

1 1 2 3 1 3 5 6 1 5 1

1 2 6 2 2 4 3 6 1 6 2

2 3 3 3 5 2 4 5 2 5 2

Re: Calculating Min and Max

Kc2 — Thu, 20 Jun 2019 15:17:01 GMT

Hi,

the values under var2 through var6 in the datalines and in the output do not match.

In the datalines suject1 var3=1 but in the output var3=2. Assuming the datalines are correct try this code:

data have;
input person_id event_id var2 var3 var4 var5 var6;
datalines;
1 1 2 3 1 3 5
1 2 6 1 2 4 1
2 3 3 1 4 2 5

run;

Proc sort data= have; by person_id event_id ; run;

proc transpose data=have out=tr_have;
by person_id event_id ;
var var2 var3 var4 var5 var6 ;
run;

proc sql;

create table wantp as
select distinct(person_id), min(col1) as minp, max(col1) as maxp
from tr_have
group by person_id;

create table wante as
select distinct(event_id),person_id, min(col1) as mine, max(col1) as maxe
from tr_have as a
group by event_id;

create table want as
select a.*,b.mine,b.maxe
from wantp as a left join wante as b
on a.person_id=b.person_id;

quit;