Hi,
Does anyone know a way to "fill out" a dataset with values from a second dataset using merge or set or other simple statements? For example, I might have one dataset as:
letter num
a .
b .
c .
And a second datset:
num
1
2
3
I need the resulting dataset to look like:
letter num
a 1
a 2
a 3
b 1
b 2
b 3
c 1
c 2
c 3
I really don't want to use an array. Thanks for any ideas!
Jon
proc sql;
create table want as
select letter,set2.num
from set1,set2;
quit;
proc sql;
create table ab as
select letter, coalesce(a.num, b.num)
from a,b;
select * from ab;
quit;
I really think the sql-version is an option here. But this might work as well, if the number of observations to fill out is fixed:
Data A;
Input Letter $ @@;
Call Missing (Num);
Datalines;
a b c
;
Run;
Data B;
Input Num @@;
Datalines;
1 2 3
;
Run;
Data Want;
Set A;
By Letter;
If First.Letter Then Do;
i=0;
Set B Nobs=MaxNum;
Do Until (i=MaxNum);
i=i+1;
Set B Point=i;
Output;
End;
End;
Run;
Hello,
data master;
input letter $ num;
datalines;
a .
b .
c .
x .
y .
z .
;
data trans;
input num;
datalines;
1
2
3
;
data want;
do until(ltrans);
i+1;
set trans end=ltrans;
call symput('var'||put(i,1.), put(num, 1.));
end;
do until(lmaster);
j+1;
set master end=lmaster;
if mod(j,3)=1 then num=symget('var1');
else if mod(j,3)=2 then num=symget('var2');
else if mod(j,3)=0 then num=symget('var3');
output;
end;
stop;
drop i j;
run;
Output
Available on demand!
Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.