Hi all,

I have a single cube with a single measure (count), I also have several
dimensions.

I need to calculate the mean of the total count for a dimension by another
dimension.

I am unsure how to go about calculating a count of my dimension.

I am using SQL2005.

Thanks for your help.

Hi,

First define your dimension set, here I am using dim.members for simplicity.
then mean can be calcualted as:

sum(dim.members)/dim.members.count

"ashvsaod" wrote:

Thanks.

I am having issues with this.

The count of the members of my dimension changes depending upon another
dimensions selection. This second dimension is used as in a where
statement.

Essentially Dimension 1 contains the items purchased
Dimension 2 is how many times the customer has been to our store

Thus the user will select from dimension2 a number as an example lets say 2.
This will filter the data to all customers that have been to our store
twice. How do I then allow the mean to calculate the number of members of
dimension 1 correctly?


"yongli" <yongli (AT) discussions (DOT) microsoft.com> wrote

If you don't want the dimension1 count vary by other dimension, you could
force allmember for other dimension. For example

count(dimension1.members, dimension2.allmember)

"ashvsaod" wrote: