remove first of the duplicate

[Smitha9]

Hi,

I have a dataset:

ID      set

A        1

A        2

A        1

B        2

B        1

C        2

I want to remove the only first set of duplicate.

I want:

A    2

B    2

C    2

Can I do that in SAS?

thank you in advance.

[Reeza]

Assuming first means lowest? What happens if a record has duplicates in the 2s?

proc sort data=have;
by ID set;
run;

data want;
set have;
by ID set;
retain counter 0;
if first.id then counter+1;
if counter=2;
run;

---

@Smitha9 wrote:

Hi,

I have a dataset:

ID      set

A        1

A        2

A        1

B        2

B        1

C        2

I want to remove the only first set of duplicate.

I want:

A    2

B    2

C    2

Can I do that in SAS?

thank you in advance.

---

[Kurt_Bremser]

The way you handle A, it looks rather as if you want to keep only the last (largest) value?

[ballardw]

Suppose you  have a dataset:

ID set

A 1

A 2

A 3

A 1

A 2

A 2

A 3

B 2

B 1

C 2

What would the result look like? Your "rule" might be a bit understated. Lists of output with some description of rules and be somewhat misleading. Now there are two different duplicates for A and a non-duplicated version. So, what is the rule?