Okay let me alter that procedure slightly.


1. Data sorted is OK. (Good)
2. Starting at the BOTTOM of the dataset we check the first column M value. should check rows even if its not at the end of a quarter, because this is not ALWAYS a requirement. sorry for the confusion
3. Check the row above if the column M value is the same and the end of month prior, check the next row above and see if it's same value and end of 2nd month prior. If yes, then move on
If either row is NOT the same value and/or month(s) prior, copy upward to get them
"Each time no more than two months are being added, we're not going backward in time indefinitely, just the current quarter." Is there anyway we could keep going until: first check M value. If its not the same don't keep going. If it is the same, check the date. Keep going until you reach that date (but don't overwrite). I am pretty sure this will entail never having more than two months, but I will need to verify my dataset. Seems better to have it this way though if its possible, to be sure.

To clarify the main goal, its basically to fill in all of the months without overwriting any data thats already there. (But not every ID will have the same start end end points. I just want to fill in the data between quarters)

Does that make sense or should I clarify?

Thanks,
Michael