I need to convert a Moving Average Trend calculation from BW into Datasphere. Purpose is to calculate a Rolling 12 months for each month in a complete year (e.g. DEC2023 must show data from JAN2023-DEC2023, NOV2023 must show data from DEC2022-NOV2023 etc).
This is calculated in BW via a replacement path with the "Current Member" functionality, which results in the correct information without multiplying the amount of records.
I was able to create the same calculation in Datasphere, where I start by creating a view that maps one month (e.g. DEC2023) with all necessary months to become the same calculation (being JAN2023-DEC2023). After joining this temperarily view to the original data view, the result is correct but the amount of records is multiplied by 12 due to these month mapping. It is also not possible to use an aggregation operator, because every month in the data view is linked to 12 months for the calculation.
My question: is there another way to implement this Moving Average Calculation to avoid the multiplication of the number of records?
Kind regards,
Lieven
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.