Hi guys, anyone can help me to simplify my code in the way to format my final dataset want?...maybe using loops rather than write down by hand the 4 combinations (class/type)?
data new;
input name class $ type $;
datalines;
1 a yes
2 a no
3 a yes
4 b no
5 b yes
6 a yes
7 b no
8 b no
9 b yes
10 a yes
11 a yes
;
run;
*creating macrovariables;
proc sql noprint;
select count(distinct name) into: safea
from new
where class='a';
quit;
proc sql noprint;
select count(distinct name) into: safeb
from new
where class='b';
quit;
%put &safea;
%put &safeb;
proc sql noprint;
create table test as
select distinct class, type, count(distinct name) as n
from new
group by class,type;
quit;
*creating the format of my want dataset;
data want;
set test;
if class='a' and type='yes' then nc=put(n,best12.)||'('||put(n/&safea*100,5.1)||')';
if class='a' and type='no' then nc=put(n,best12.)||'('||put(n/&safea*100,5.1)||')';
if class='b' and type='yes' then nc=put(n,best12.)||'('||put(n/&safeb*100,5.1)||')';
if class='b' and type='no' then nc=put(n,best12.)||'('||put(n/&safeb*100,5.1)||')';
drop n;
run;
Thanks
V.
You could do this more compactly as folows :
proc sql;
create table want as
select
class,
type,
catt(ncCount, " (", put(100*ncCount/sum(ncCount), 5.1), ")") as nc
from (select class, type, count(*) as ncCount from new group by class, type)
group by class;
quit;
PG
You could do this more compactly as folows :
proc sql;
create table want as
select
class,
type,
catt(ncCount, " (", put(100*ncCount/sum(ncCount), 5.1), ")") as nc
from (select class, type, count(*) as ncCount from new group by class, type)
group by class;
quit;
PG
I prefer using procedures, such as proc freq.
proc sort data=new; by class type;
proc freq data=new noprint;
by class;
table type/out=want1 totpct;
run;
data want2;
set want1;
nc=cat("", trim(strip(put(count, 8.))), "(",trim(strip(put(percent/100, percent8.1))), ")");
run;
Just Brilliant, Thank you.
It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.
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.
Ready to level-up your skills? Choose your own adventure.