selection data

[raveena]

Hi,

I have a dataset with columns clam,lineno and paiddate..Each clamno no has multiple lineno with paiddate.
For example if am processing for june month..

selection criteria:
1.Need to should skip the clam 1 bcoz the clamno is not within the processing month.
2.I can select the clam 2 bcoz the clamno is within the processing month.
3.Need to skip the clam 3 bcoz its ot within the processing month.

clam      lineno         paiddate
1          001         04/11/2011
           002         06/15/2011
           003         07/02/2011

2          001         04/11/2011
           004         06/15/2011
           005         06/28/2011

3          002         04/11/2011
           006         05/11/2011
           007         05/28/2011

and so on..........

So output should be ,

clam      lineno         paiddate

2          001         04/11/2011
           004         06/15/2011
           005         06/28/2011

Please let me know how to get this desired output.

Thanks

[NN]

Hi..

A simple method would be to select * and max of paiddate

group the data by clam. and then filter out all records where month of the new column is june .

[art297]

One way of doing it might be:

data need;

  set have;

  retain claim;

  if not missing(clam) then claim=clam;

run;

proc sql noprint;

  create table want (drop=claim) as

    select * from need

      group by claim

        having month(max(paiddate)) eq 6

          order by lineno

;

quit;