How to extract observations who have information for each month

[viollete]

Hi ,

I have this information (this is just an example):

Year Month Hospital Nr_beds

2016 12 RA 210

2017 1 RA 215

2017 2 RA .

2016 12 RB .

2017 1 RB 650

2017 2 RB 651

2016 12 RC 120

2017 1 RC 150

2017 2 RC 170

I want to extract only hospital who has no missing values in the variable Nr_beds. In new table I want to have just hospital RC, but not RA (cause it does not have information for time 2017/2) and hospital RB (cause it does not have information for time 2016/12)

[RW9]

Post test data in the form of a datastep.

A simple where seems to cover your request:

data want;
  set have;
  where hospital="RC" and nr_beds ne .;
run;

[Ksharp]

data have;
input Year Month Hospital $ Nr_beds;
cards;
2016 12 RA 210
2017 1 RA 215
2017 2 RA .
2016 12 RB .
2017 1 RB 650
2017 2 RB 651
2016 12 RC 120
2017 1 RC 150
2017 2 RC 170
;

proc sql;
create table want as
 select * from have
  group by Hospital
   having nmiss(Nr_beds)=0;
quit;

[Ksharp]

data have;
input Year Month Hospital $ Nr_beds;
cards;
2016 12 RA 210
2017 1 RA 215
2017 2 RA .
2016 12 RB .
2017 1 RB 650
2017 2 RB 651
2016 12 RC 120
2017 1 RC 150
2017 2 RC 170
;

proc sql;
create table want as
 select * from have
  group by Hospital
   having nmiss(Nr_beds)=0;
quit;