Hi Dipen,
I managed to figure out the reason meanwhile. There was double "initial stock" (=record type 1) with two different values for 0CALDAY in the cube we were trying to compress.
As 0CALDAY is the special aggregation characteristic controlling the whole temporal stock calculation, this is of course not valid, because you need values for exactly one day to give your calculation basis (record type 1.) The rest of the data is split to "historical" (record type 2) and "newer than the initial stock" (record type 0).
The fact that we ended up with the double data was caused by a "delete PSA data" process not deleting the data but still giving a green status - but that is again a different cup of tea.
Check your data to see if probably yours is also "double", in HANA Studio it should be fairly easy to do a select with group by calday to check if there is only one distinct value.
Oliver