compare rows in dataset and deleting

[Nrjn7]

ID	Event	Code	value
1	A
1	B	1	6
1	B	1	7
1	B	1	8
2	A	5	7
2	B	6	8

want

ID	Event	Code	Value
1	A
1	B	1	6
2	A	5	7
2	B	6	8

This is just an example, but for my dataset, I need to be able to compare ID and code.

If the ID and code are the same, then I want to only retain one row and delete all the other rows.

Can you please suggest?

[ballardw · Accepted solution]

Easiest might be

Proc sort data=have out=want nodupkey;

   by Id code;

run;

Create a new output data set Want so you don't accidentally delete things you want. This way your original data set remains.

If EVENT plays any role in this process, which you did not mention, then you may want to sort by ID Event Code; Maybe.