I have a cube with three main dimensions (time, product, distributor) and a
measure that's an aggregated average of two values in the fact table ( in
SQL syntax it's... sum ([amount] * [number of observations]) / [number of
observations] ). I'm trying to create a utility dimension that would group
the averages into range buckets 0, 0-5,5-10, 10-15, etc., but I want it to
work for every possible selection in the cube. I've tried creating another
column in the fact table to contain a key value of each possible range, but
this is too limited - not dynamic enough for all the possible ways to slice
the data. I've tried using custom member formulas and also an analysis
dimension with calculated members, but neither get me close. It's almost as
though I need to create a dimension table that contains the ranges, but
instead of joining on a key in the fact table with SQL, I need to join on
the value of a custom MDX expression.

Hope this makes sense. Thanks for your help.

-Keith

You could create a so-called calculation dimension. You can have dynamic
threshholds, but can't have a dynamic # of buckets. Here's a generic
example:

http://www.tomchester.net/articlesdo...dimension.html

tom @ the domain below
www.tomchester.net


"Keith" <x@x.com> wrote

Thanks Tom. Checked out your site and your example. Both are great. When I
mentioned "analysis dimension" I guess I meant "calculation dimension". I
had learned the technique from one of Spofford's posts. I'll rethink the
problem. Maybe I can use that after all. Thanks for your help.

-Keith

"Tom Chester" <publicNOSPAM (AT) tomchester (DOT) net> wrote