Eliminate rows if more than one

[Q1983]

data have;

input id Cd $ flg $ DATE1 DATE9. ;

format date1 date9.;

datalines;

11 chg b 13apr2019

11 chg i 11apr2019

12 chg e 6may2017

13 chg b 5jun2019

14 add d 4mar2019

14 chg y 1jan2019

;run;

I want to eliminate all rows where the number of rows per id is more than 1.  So In this example I would eliminate 11 and 14 and only keep 12 and 13.  Using a proc sort nodupkey would not work in this case.

[Tom]

So only keep the IDs that do NOT have multiple rows?

data want;
  set have;
  by id;
  if first.id and last.id;
run;

[novinosrin]

data have;

input id Cd $ flg $ DATE1 DATE9. ;

format date1 date9.;

datalines;
11 chg b 13apr2019
11 chg i 11apr2019
12 chg e 6may2017
13 chg b 5jun2019
14 add d 4mar2019
14 chg y 1jan2019
;

proc sort data=have nouniquekeys uniqueout=singles ;
by id;
run;

[Reeza]

You want the NOUNIQUEKEY option with the UNIQUEOUT option within PROC SORT.

proc sort data=have nouniquekey uniqueout=want;
by id;
run;