I've created a cube which has a measure "Balance
Outstanding" as of a certain date. And the cube has these
balances 'as at' each day of the year.

It's fine to sum this across the Debtor dimension, and the
State dimension and the type-of-debt dimension etc etc.
But it's meaningless against the time dimension.

I can find numerous articles that tell me that it's semi-
additive; that it's a snapshot like a stock count as at a
certain date; that I need to 'disable the aggregate over
time'.

But no bol or on-line article that I can find tells me how
to do this. Can anyone explain it to me please?

Thanks

Geoff

Hi Geoff,
The way you do this is by creating a calculated member
with the definition like :
(Measures.<<Measure Name>>, ClosingPeriod
(<<Dimension>>.<<Level>>))

"ClosingPeriod" fetches the last member in the specified
dimension at the specified Level. So the tuple with the
Measure alongwith ClosingPeriod function will give you the
snapshot value at the level you need.
So if BalanceOutstanding is the measure and Day is the
name of the level in your time dimension you wish to have
the snapshot value for, the definition would look like
([Measures].[BalanceOutstanding],ClosingPeriod([Time].
[Day]))

You don't need to make any changes to your fact table or
change any aggregation in the cube.

Hope this helps! Let me know if you need more help.
- Priya