Changing records by id and occurrence

[Sean_OConnor]

Folks,

I'm having an issue with something at the moment which I'd be grateful for some assistance on.  

I've a dataset at the moment which has a number of 0 for variable occurrence.

What I would like to do is change where there is a 0 to 1 and subsequently change the 1 to 2 and so on. Below gives a better idea.

Data in current format;

ID	Occurrence	Value
63	0	45
63	1	65
63	2	98
22	1	100
23	1	125
24	1	130
25	1	130
45	1	96
45	2	88
45	3	44

What I would like to occur;

ID	Occurrence	Value
63	1	45
63	2	65
63	3	98
22	1	100
23	1	125
24	1	130
25	1	130
45	1	96
45	2	88
45	3	44

[Astounding]

If there is an instance of occurrence=0, will it always be for the first observation for an ID?

Here's an approach that makes a reasonable guess about how to handle that question:

data want;

set have;

by ID;

if first.ID then increment=0;

retain increment;

if occurrence=0 then increment=1;

occurrence = occurrence + increment;

run;