# Calculated member on time dimension with conditions

Greetings,
I have a calculated member for my time dimension that gives me all the date members between the last day of last month and 6 months before that date:

``````Aggregate(
[Time].[Month].DateMembersBetween('7 months ago', 'one day before first day of current month')
)
``````

Now, I would like to have this for the first two weeks of the month, and to include current month if the date of today is > 15th, so what I did is:

```````CASE
WHEN
DateDiffDays('first day of current month', now()) > 15
THEN
Aggregate(
[Time].[Month].DateMembersBetween('6 months ago', 'today')
)
ELSE
Aggregate(
[Time].[Month].DateMembersBetween('7 months ago', 'one day before first day of current month')
)
END
`
``````

but it doesn’t seem to go anywhere, I don’t get any time member when I try to use this calculated member.
Any idea? I tried to build a measure with the check I made in the first when clause, and it seems to work just fine.

While there is little difference in the construction sequence for calculated measures, the construction sequence does make a difference for the calculated member.
To keep the dimension member aggregated, you should raise the `Aggregate` function to the top and bring the conditional selection to the lowest level.
The same words placed in a different order provide the expected result.

``````Aggregate(
[Time].[Month].DateMembersBetween(
--expression for start date
CASE WHEN DateDiffDays('first day of current month', now()) > 15
THEN
'6 months ago'
ELSE
'7 months ago'
END,
--expression for end date
CASE WHEN DateDiffDays('first day of current month', now()) > 15
THEN
'today'
ELSE
'one day before first day of current month'
END)
)
``````

Regards,
Oskars / support@eazybi.com

thanks very much @oskars.laganovskis!
What’s the cause of this behaviour?
regards

While the aggregation within a measure passes on a single numeric value, the calculated members within dimensions are aggregated sets of members. Therefore - the calculated member should have the Aggregation on the top.