I'm connecting Power BI "live" to an SSAS data model, therefore I can't add calculated columns (which otherwise would be the simplest solution, I guess). The table below shows a list of dates, each row is a particular service provided to a client, the "Service Referral Date" and "Service Start Date" are pulled from the "Service" fact table.
Since I can't add calculated columns, I defined two measures (which I hoped would just do the same) as follows:
Add To Referral Date = dateadd('FACT Services'[Service Referral Date],6,MONTH)
Add To Start Date = dateadd('FACT Services'[Service Start Date],6,MONTH)
As shown in the table, this approach works, but not for all dates.
I did some research I and found that DateAdd only works with contiguous dates, so somehow I'm suspecting that the problem is caused by "missing dates" in the fact table because the problem seems to exist only when the calculated date is bigger than today, and such dates do not exist in the fact table. I'm not quite sure how to approach that.
Anyway, all I'm trying to do is to add x number of months to a given date. Would there be a more straight forward solution than "DateAdd"?
(And if a more "nerdy" solution is required: The data model contains for both dates an independently linked "Date" table which contains all required dates, i. e. the Referral Date is linked to table "Referral Dates", and the Start Date is linked to table "Start Dates", but I'm not sure how to reference the dates there from within the measure).
Thank you very much for putting me on the right track. I didn't know about edate, and it works perfectly, I just needed the following "workaround":
Edate didn't allow me to reference to any of the existing dates (Start, Referral), it only allowed me to reference another measure. So I set up a measure "MaxReferralDate" and "MaxStartDate", and then I added 6 months via edate(MaxReferralDate, 6).
(I have a sense that it might be a bit of a "brute force" approach, but as I'm digging deeper into filter and row context, maybe I'll find an (even) more elegant solution.)
But this absolutely works for my purpose, so thank you very much!