01-22-2018 01:50 PM
I am looking for some help on creating a macro that will identify continuous enrollment with a given gap. For example, we will allow a month gap and still allow continuous enrollment. The data structure I have is....
Memb_id enroll_yrmo (number - not date)
I have done this before but it was with from and thru dates with a single record per member. Now, I have a record for every month and it isn't even a date. It is a number. I would love your help and input.
01-22-2018 02:02 PM
Here's one approach, assuming your data set is already sorted:
dif_date = dif(enroll_yrmo);
month = mod(enroll_yrmo, 100);
if first.memb_id then do;
if month > 1 and dif_date=1 then return;
if month=1 and dif_date=89 then return;
group + 1;
Since you described the result, but didn't illustrate it, here's what this is doing. It assumes that a "gap" means that a year/month is missing from the sequence (for the same value of MEMB_ID). It creates a new variable (GROUP) that is 1 for the first set of sequential observations, but increases whenever a gap is found.