Hi,

We are trying to build a cube with has time categorized as Year -
Quarter - Month. The quarterly value need not be an aggregated value
of all the months in that quarter put together. We are trying to build
the cube using AS 2000. When we choose the month entry, the right data
gets displayed but the proble occurs when we choose the quarter. It
gives the aggredated value of all the months under that quarter.

Is there a way to suppress this behaviour?

Any pointers are appreciated.

Thanks

It depends on what value you want to appear at the quarter level and
what edition of AS you are using.

Custom Member formulas or Custom Member formulas are probably what you
are after. You can find plenty of information and some examples in Books
On Line.

Unary Operators might be an option, but are more often used in Financial
type dimensions.

If you are using enterprise edition you would also have the option of
using Calculated Cells, you can use quite sophisticated formulas in
these, but you do need to be careful not to over use them as they can
have a negative impact on performance.

--
Regards
Darren Gosbell - SQL Server MVP
Blog: http://www.geekswithblogs.net/darrengosbell

In article <1158068256.887011.256680 (AT) m73g2000cwd (DOT) googlegroups.com>,
bharathram.c (AT) gmail (DOT) com says...

You should have a look at the Custom Rollup Formula in the dimension editor.

There you can define an MDX expression to be used for the agregation at each
level. It should look like this for the time dimension

iif(
isleaf(time.currentmember) //month
,
CalculationPassValue(..,0) // reads the agregated data at month level
,
iff(time.currentmember,level.ordinal=2 // quarter
,
max(time.currentmember.siblings) // max of months or whatever
custom agreg.
,
sum(descendands(time.currentmember,time.month,LEAV ES) // year =
sum(months)
)
)

The sintax is not 100% correct.

regards,
Radu

<bharathram.c (AT) gmail (DOT) com> wrote