https://communities.sas.com/t5/SAS-Procedures/Proc-sort-Nodupkey-but-I-want-to-keep-the-first-occurance/m-p/57992#M16190 Another option if you want more "control" is to consider using a DATA step with a SET statement, a BY statement (input file must be sorted by the variable list), and use IF FIRST.variable_name and if necessary IF LAST.variable_name, and the OUTPUT statement to create the file as needed.<BR /> <BR /> Scott Barry<BR /> SBBWorks, Inc. Thu, 30 Oct 2008 02:34:09 GMT sbb 2008-10-30T02:34:09Z https://communities.sas.com/t5/SAS-Procedures/Proc-sort-Nodupkey-but-I-want-to-keep-the-first-occurance/m-p/57990#M16188 I have a data set in which I have duplicates (rcode, pcodes, and prov). I want to create a new table in which these are deduped, but retaining the first time they were experienced as unique by (keydt). I cant use proc sort nodupkey because I need to include keydate in the sort, and will yield no dups by that particular date. If I don’t use the keydate the sort will then sort by rcode, pcode and prov; and will not yield the first time the unique combination was observed.<BR /> <BR /> <BR /> data xx;<BR /> format keydt date. rcode $4. pcode $5.;<BR /> input ID keydt :date. rcode pcode pr;<BR /> cards;<BR /> 8001 20-Jan-06 634 J0886 26<BR /> 8001 20-Jan-06 636 J1270 26<BR /> 8001 20-Jan-06 821 90999 26<BR /> 8001 06-Feb-06 634 J0886 26<BR /> 8001 06-Feb-06 636 J1270 26<BR /> 8001 06-Feb-06 821 90999 26<BR /> 8001 23-Feb-06 634 J0886 26<BR /> 8001 23-Feb-06 636 J1270 26<BR /> 8001 23-Feb-06 821 90935 26<BR /> 8001 03-Mar-06 634 J0886 26<BR /> 8001 03-Mar-06 636 J1270 26<BR /> 8001 03-Mar-06 821 90935 26<BR /> 8001 10-Apr-06 636 J1270 26<BR /> 8001 10-Apr-06 821 90935 26<BR /> 8001 20-Apr-06 634 J0886 26<BR /> 8001 20-Apr-06 636 J1270 26<BR /> 8001 20-Apr-06 821 90935 26<BR /> 8001 13-May-06 270 A4657 26<BR /> 8001 13-May-06 634 J0886 26<BR /> 8001 13-May-06 636 J1270 26<BR /> 8001 13-May-06 821 90935 26<BR /> 8001 16-May-06 270 A4657 26<BR /> 8001 16-May-06 634 J0886 26<BR /> 8001 16-May-06 636 J1270 26<BR /> 8001 16-May-06 821 90935 26<BR /> 8001 22-May-06 634 J0886 26<BR /> 8001 22-May-06 636 J1270 26<BR /> 8001 22-May-06 821 90935 26<BR /> 8001 07-Jun-06 634 J0886 26<BR /> 8001 07-Jun-06 636 J1270 26<BR /> 8001 07-Jun-06 821 90935 26<BR /> 8001 20-Jun-06 634 J0886 26<BR /> 8001 20-Jun-06 636 J1270 26<BR /> 8001 20-Jun-06 821 90935 26<BR /> 8001 07-Jul-06 270 A4657 26<BR /> 8001 07-Jul-06 634 J0886 26<BR /> 8001 07-Jul-06 636 J1270 26<BR /> 8001 07-Jul-06 821 90935 26<BR /> 8001 14-Jul-06 634 J0886 26<BR /> <BR /> ;<BR /> run; Wed, 29 Oct 2008 22:56:52 GMT https://communities.sas.com/t5/SAS-Procedures/Proc-sort-Nodupkey-but-I-want-to-keep-the-first-occurance/m-p/57990#M16188 Ray_H 2008-10-29T22:56:52Z https://communities.sas.com/t5/SAS-Procedures/Proc-sort-Nodupkey-but-I-want-to-keep-the-first-occurance/m-p/57991#M16189 Consider using multiple sorts, first without and then with the NODUPSKEY and also including the EQUALS parameter for your BY key variables. <BR /> <BR /> Scott Barry<BR /> SBBWorks, Inc. Thu, 30 Oct 2008 02:18:21 GMT https://communities.sas.com/t5/SAS-Procedures/Proc-sort-Nodupkey-but-I-want-to-keep-the-first-occurance/m-p/57991#M16189 sbb 2008-10-30T02:18:21Z https://communities.sas.com/t5/SAS-Procedures/Proc-sort-Nodupkey-but-I-want-to-keep-the-first-occurance/m-p/57992#M16190 Another option if you want more "control" is to consider using a DATA step with a SET statement, a BY statement (input file must be sorted by the variable list), and use IF FIRST.variable_name and if necessary IF LAST.variable_name, and the OUTPUT statement to create the file as needed.<BR /> <BR /> Scott Barry<BR /> SBBWorks, Inc. Thu, 30 Oct 2008 02:34:09 GMT https://communities.sas.com/t5/SAS-Procedures/Proc-sort-Nodupkey-but-I-want-to-keep-the-first-occurance/m-p/57992#M16190 sbb 2008-10-30T02:34:09Z