Date: Fri, 18 Apr 2008 20:37:11 +0300
Reply-To: Mugu Gai Pan <mugu@PCN.NO>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Mugu Gai Pan <mugu@PCN.NO>
Subject: little help on Proc Sql and frequency
Content-Type: text/plain; format=flowed; charset="iso-8859-1";
reply-type=original
say I have the following raw data set
data test; input vin_nbr $10. @12 code $2.;
datalines;
0911221122 j7
0911221122 j7
0911221122 j7
0911221122 j1
0311221122 j4
0311221122 j4
0311221122 j4
0311221122 j5
0311221122 j1
;
run;
my goal is to report total number of vin_nbr that match the following
conditions
1. total number of distinct vin_nbr
2. total number of vin_nbr which have a code "j1"
proc sql;
create table result as
select count(*) as total from(select distinct vin_nbr from test);
create table result2 as
select count(*) as total from(select distinct vin_nbr from test
where code in('j1'));
quit;
data _null_;set result ;put(_all_)(=);run;
data _null_;set result2 ;put(_all_)(=);run;
how would I count total number of all codes within the total vin_nbr?
ie. j7 occurred 3 times but only within 1 vin_nbr, j1 2 times within 2
vin_nbr