Hi,
In my data, I have multiple rows for one person and I'd like to combine all the rows into one unique row while summing the values. For example:
Where year_id and bat_id are equal, I would like to combine the rows and sum the column values. So for 2015, I would just want one row with values:
82 for G, 232 for AB, and so one. What procedure is best able to do this, or could it be done in the data step?
Thank you,
Justin
you could try the proc means
data have;
input year_id bat_id$ g ab r h _2b;
cards;
2015 almoa001 51 178 30 47 9
2015 almoa001 31 54 6 11 3
2014 almoa001 32 98 9 26 5
2014 almoa001 27 106 10 21 5
;
proc means data=have nway;
class year_id bat_id;
var g ab r h _2b;
output out=sums sum=/autoname;
run;
you could try the proc means
data have;
input year_id bat_id$ g ab r h _2b;
cards;
2015 almoa001 51 178 30 47 9
2015 almoa001 31 54 6 11 3
2014 almoa001 32 98 9 26 5
2014 almoa001 27 106 10 21 5
;
proc means data=have nway;
class year_id bat_id;
var g ab r h _2b;
output out=sums sum=/autoname;
run;
alternatively proc sql
proc sql;
create table want as select year_id, bat_id,sum(g) as g, sum(ab) as ab, sum(r) as r, sum(h) as h, sum(_2b) as _2b
from have group by year_id,bat_id;
quit;
If you use @Jagadishkatam's Proc Means solution with Sum= instead of Sum= / autoname then the variable names would stay unchanged if that would be the preferred behavior.
Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!
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.