Group by count obs in proc sq;

[BrahmanandaRao]

data dup;
set sashelp.class;
proc sort data =dup;
by age;
run;

data d;
set dup;
by Age ;
if first.Age then n=0;
n+1;proc print;
run;

Hi Guys

Good Morning 

How to get same output using proc sql

[PeterClemmensen]

You want an accumulating sum in SQL, correct?

[Kurt_Bremser]

Very simple.

COUNT(*) GROUP BY AGE in the SQL SELECT

Edit:

Belay that, you want a running count.

Maxim 14: Use the Right Tool, which is the data step. Don't waste your brain cycles on stupidity when there's a simple solution you already have.

[BrahmanandaRao]

Hi Krutbremser
i want count as like as datastep output in proc sql
your code give only count age wise but i want

--------------------------------------------------------------------------------
The SAS System


Obs Name Sex Age Height Weight n
1 Joyce F 11 51.3 50.5 1
2 Thomas M 11 57.5 85.0 2
3 James M 12 57.3 83.0 1
4 Jane F 12 59.8 84.5 2
5 John M 12 59.0 99.5 3
6 Louise F 12 56.3 77.0 4
7 Robert M 12 64.8 128.0 5
8 Alice F 13 56.5 84.0 1
9 Barbara F 13 65.3 98.0 2
10 Jeffrey M 13 62.5 84.0 3
11 Alfred M 14 69.0 112.5 1
12 Carol F 14 62.8 102.5 2
13 Henry M 14 63.5 102.5 3
14 Judy F 14 64.3 90.0 4
15 Janet F 15 62.5 112.5 1
16 Mary F 15 66.5 112.0 2
17 Ronald M 15 67.0 133.0 3
18 William M 15 66.5 112.0 4
19 Philip M 16 72.0 150.0 1

[Kurt_Bremser · Accepted solution]

Just to show how it's done:

proc sql;
select
  *,
  (
    select count(*)
    from sashelp.class t2
    where t2.age = t1.age
    and t2.name le t1.name
  ) as n
from sashelp.class t1
order by age, name
;
quit;

Note that sub-selects perform horribly, so on any larger dataset this will cause time and resource issues.

To illustrate this:

data class;
set sashelp.class;
do i = 1 to 100000;
  output;
end;
run;

proc sql;
create table want as
  select
    *,
    (
      select count(*)
      from class t2
      where t2.age = t1.age
      and t2.name le t1.name
    ) as n
  from class t1
  order by age, name
;
quit;

proc sort
  data=class
  out=dup
;
by age;
run;

data d;
set dup;
by Age;
if first.Age
then n = 1;
else n + 1;
run;

Log:

 73         data class;
 74         set sashelp.class;
 75         do i = 1 to 100000;
 76           output;
 77         end;
 78         run;
 
 NOTE: There were 19 observations read from the data set SASHELP.CLASS.
 NOTE: The data set WORK.CLASS has 1900000 observations and 6 variables.
 NOTE:  Verwendet wurde: DATA statement - (Gesamtverarbeitungszeit):
       real time           0.10 seconds
       cpu time            0.09 seconds
       
 
 79         
 80         proc sql;
 81         create table want as
 82           select
 83             *,
 84             (
 85               select count(*)
 86               from class t2
 87               where t2.age = t1.age
 88               and t2.name le t1.name
 89             ) as n
 90           from class t1
 91           order by age, name
 92         ;
 NOTE: Table WORK.WANT created, with 1900000 rows and 7 columns.
 
 93         quit;
 NOTE:  Verwendet wurde: PROZEDUR SQL - (Gesamtverarbeitungszeit):
       real time           5.82 seconds
       cpu time            5.97 seconds
       
 
 94         
 95         proc sort
 96           data=class
 97           out=dup
 98         ;
 99         by age;
 100        run;
 
 NOTE: There were 1900000 observations read from the data set WORK.CLASS.
 NOTE: The data set WORK.DUP has 1900000 observations and 6 variables.
 NOTE:  Verwendet wurde: PROZEDUR SORT - (Gesamtverarbeitungszeit):
       real time           0.51 seconds
       cpu time            0.88 seconds
       
 
 101        
 102        data d;
 103        set dup;
 104        by Age;
 105        if first.Age
 106        then n = 1;
 107        else n + 1;
 108        run;
 
 NOTE: There were 1900000 observations read from the data set WORK.DUP.
 NOTE: The data set WORK.D has 1900000 observations and 7 variables.
 NOTE:  Verwendet wurde: DATA statement - (Gesamtverarbeitungszeit):
       real time           0.31 seconds
       cpu time            0.30 seconds

You can see that there's already a factor of 6 between DATA and SQL; this difference grow exponentially with dataset sizes.

Rule of thumb: in 90% of cases, DATA step solutions outperform SQL. On top of being easier to read and maintain.