Looping over to cumulate scores in SAS - sas

Question

Given this dataset in SAS, i would like to calculate:
the total number of wins obtained for each team
using do loop to identify how many matches has been performed
data array;
infile datalines dlm=',' firstobs=2;
datalines;
game,winning_team, points1, loosing_team, points2
May2, Berfield, 12, Jacksons, 10
June3, Jacksons, 23, North, 22
June5, UCI, 12, Jacksons, 10
June23, Nottingham, 12, Jr High, 11
;
run;
May i know how should i start? i used the following code but still can't work
array ray(1) game;
do i=1 to dim(game);
i=i+1;
output;
end;

Answer 1

If you want the number of wins per team it looks like you could just get the frequency counts for variable WINING_TEAM. The total count will be the number of games played. No looping is required.
proc freq data=array ;
tables winning_team;
run;
PS ARRAY is a strange name for a dataset. A more descriptive name might be RESULTS since the data appears to list the results of games played.

Answer 2

A win/loss ratio for each team can be computed by FREQ if you first pivot each row into two categorical observations.
Example:
data season;
infile datalines dlm=',' ;
length game $10 winning_team $20 points1 8 losing_team $20 points2 8;
input game winning_team points1 losing_team points2;
datalines;
May2, Berfield, 12, Jacksons, 10
June3, Jacksons, 23, North, 22
June5, UCI, 12, Jacksons, 10
June23, Nottingham, 12, Jr High, 11
;
run;
proc transpose data=season out=winloss(keep=_name_ col1);
by _all_ notsorted;
var winning_team losing_team;
run;
proc datasets nolist lib=work;
modify winloss;
rename _name_ = status col1=team;
label status = ' ';
run;
proc sort data=winloss;
by team;
proc freq noprint data=winloss;
by team;
table status / out=freqs;
run;

Answer 3

This will give you those two pieces of information:
proc sql;
select
count(*) as total_number_of_games
from array;
select
winning_team as team
,count(*) as number_of_wins
from array
group by 1;
quit;
You don't need to use a do loop as the total number of games played will simply be the number of rows in the table.