considering a row with missing data via sql

[michtka]

x
Hi guys, I would like to obtain the next table total:

N       TRT     col0

1        2        Number of subjects

.        1        Number of subjects

I use proc sql to try to get this:

input subjid trt fday  tday;
    datalines;
    1  1 1 5
    2  1 . 4
    2  1 . 3
    3  1 1 4
    3  2 1 -5
    4  2 1 2
    4  1 1 4
    ;
    run;

    proc sql;
   create table total as
   select count(distinct subjid) as n, trt 'Treatment', 'number of subjects' as col0
   from new
   where fday ne . and tday le  0
   group by trt;
   quit;

    proc print data=total noobs; run;

But I obtain only one row:

N   TRT     col0

1    2        Number of subjects

Not appearing the row with missing data.

Can anyone help me to write the code via sql to consider this row with missing data?

Thanks,

V

[Ksharp]

You can't select 1 anymore  because your WHERE condition has already exclude it. But you can get it by your hand.

data new;
input subjid trt fday  tday;
    datalines;
    1  1 1 5
    2  1 . 4
    2  1 . 3
    3  1 1 4
    3  2 1 -5
    4  2 1 2
    4  1 1 4
    ;
    run;


    proc sql;
   create table total as
   select count(distinct subjid) as n, trt 'Treatment', 'number of subjects' as col0
   from new
   where fday ne . and tday le  0
   group by trt
    union 
   select .,1,'number of subjects' from new(obs=1)
;
   quit;

Ksharp

[ChrisNZ]

Only one row is read from your table.

Only row

3 2 1 -5

matches

where fday ne . and tday le  0