I know that no OLAP write back to overwrite aggregated data after aggregation is possible
in SAS OLAP*. So when the end user wants to correct aggregated data I let them do it
in a correction datatable in my oracle database. now the question is, how do I tell my sas cube to NOT use the data in the fact table IF a "corrected" record exists in the the relational datatable that holds the corrections? I guess I have to make the right data available in a SQL view (i.e. I sort it out in SQL) but I want to avoid digging throug the GB's of fact data each time. has anyone done something like this before?
I plan to have a scenario dimension:
in this case when the user selected "acutal" a corrected value would display if it
exists, if not take the value loaded against "Actual" itself.