some kind of "dynamic" dimension

Hi All,

I'd like to model a numeric value with range 0-100 in two different ways in the same cube.
Sometime it's helpful to have it as a dimension, and sometimes it's helpful to calculate the avg and have it as a measure.
When the user wants to see it as a measure the dimension will be hidden, on the other hand when the user wants to see it as a dimension the measure(avg) will be hidden.
Can I model a situation like this or do I have to use two different cubes?

Tnx in advance

nickall

Don't know what MDDB sytem you are using, but couldn't you load the same value twice? Once as a measure and once as a dimension?

ngalemmo wrote:Don't know what MDDB sytem you are using, but couldn't you load the same value twice? Once as a measure and once as a dimension?

I'm using pentaho community version.
Yes i can load the same value twice, but it doesn't make sense if i watch at the primary events.
In this case i have a measure (avg) and the dimension value that are the same thing.

But it would work at aggregate levels above the dimension, correct? Just use either the measure or the dimension in your analyisis...

Yes, it would work at the aggregate levels! but i'd like to make it work with the primary events!

What, exactly, is a primary event? Is the numeric dimension a definer of a primary event or may a primary event contain multiple numeric values? In either case, if you exclude the numeric dimension in the analysis of a primary event, the avg(measure) will calculate correctly.