exclude a group of observations in terms of one observation

[France]

Dear all,

how can exclude a group of observations in terms of one observation?

for example, in following observations,

psn_name_cited, docdb_family_id, psn_name_citing 

apple,1,pear

apple,1,apple

apple,2,orange 

apple,2,banana

apple,2, pear

juice,3, bottle

orange,3,water

I would like to exclude the following observations

apple,1,pear

apple,1,apple

by using the following codes,

PROC SQL;
   CREATE TABLE want  AS
      SELECT
       *
      FROM
        tryy
      GROUP BY psn_name_cited,docdb_family_id
      Having psn_name_cited ne psn_name_citing /* exclude self_citation */
;
QUIT;

However, only

apple,1,apple

been excluded from the sample.

Could you please give me some suggestions about this?

Thanks in advance.

[RW9]

Maybe joining psn_name_citing back on by psn_name_citing=psn_name, something like:

proc sql;
  create table want as
  select a.*
  from   have a
  left join have b
  on      a.psn_name=b.psn_name_citating
  where b.psn_name_citing="";
quit;

Not tested, post test data in the form of a datastep in future.

[Ksharp · Accepted solution]

PROC SQL;
   CREATE TABLE want  AS
      SELECT
       *
      FROM
        tryy
      GROUP BY psn_name_cited,docdb_family_id
      Having sum(psn_name_cited = psn_name_citing)=0 /* exclude self_citation */
;
QUIT;