How could I do like a "Conditional Summing". For example, I have is a cube
that measures A vendors delivery performace. I do a DateDiff from the
vendor's promise to the actual receipt date. A product promised on 1/1/2003
but doesn't get received until 1/10/2003 has a DateDiff value of 10. If it
were early say 12/25/02 they it would be -7. Well how can I some only the
values that are greater than zero (late) and create an avg for those. So
the measures is actually say when they are late, on avg they are this amount
late.

Hope I provided enough detail.

-mike

Maybe something like this...

AVG(Filter(Vendor.Vendor.Members, Measures.Diff > 0), Measures.Diff)

tom @ the domain below
www.tomchester.net


"Michael Morse" <mike.morse (AT) micromo (DOT) com> wrote

Hi,

How to do the conditional Summing.
1.Create a measure for DateDiff
now create another calculated member like below:

iif([Measures].[DateDiff]>0,avg({[Dimension]},[Measures].
[DateDiff]),null)

I think it will work.
Pls try.

Regards,
Vanchi