![]() | |
![]() |
| | Thread Tools | Display Modes |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
Hi all Here is something that I think may be a bit of a challenge... The topic of multiple hierarchies esp of a time dimension often crops up in this group. What I want to do is to be able to present my data against the accounting calendars of my customers. For example, my year starts in June whereas a particular customer's starts in April Thus, for internal reporting, I want Aug YTD to be Jun+Jul+Aug but when conducting the business review with this customer, Aug YTD becomes Apr+May+Jun+Jul+Aug ... and across all customers there are probably 12 different calendars!! I use a time analysis dimension to generate YTD against my calendar YTD = aggregate(periodstodate(time.year,time.currentmemb er),[time analysis].[current]) To follow this method, I would have to provide 12 alternate hierarchies and 12 alternate YTD calculations - very messy! An extra dimension (with 12 members) to generate these YTDs might be the way to go. But it still looks like the MDX would be tricky Any other ideas would be welcome Regards Ian Bamforth |
#3
| |||
| |||
|
|
How about using Member Property? IIF ( VAL(Customer.CurrentMember.Properties("Start")) Time.CurrentMember.Properties("Key"), Sum({Time.CurrentMember : StrToMember(Ancestor(Time.CurrentMember, Year).PrevMember.UniqueName + ".[" + Customer.CurrentMember.Properties("Start") +"]")}), Sum({StrToMember(Ancestor(Time.CurrentMember, Year).UniqueName + ".[" + Customer.CurrentMember.Properties("Start") +"]") : Time.CurrentMember} ) I have no test cube. My sample is not complete nor tested. You may have to customize it depending on the dimensional data and time hierarchy. But I hope it could help you think out better idea. Ohjoo Kwon "bammers99" <ian.bamforth (AT) gmail (DOT) com> wrote in message news:1111224005.985621.288070 (AT) z14g2000cwz (DOT) googlegroups.com... Hi all Here is something that I think may be a bit of a challenge... The topic of multiple hierarchies esp of a time dimension often crops up in this group. What I want to do is to be able to present my data against the accounting calendars of my customers. For example, my year starts in June whereas a particular customer's starts in April Thus, for internal reporting, I want Aug YTD to be Jun+Jul+Aug but when conducting the business review with this customer, Aug YTD becomes Apr+May+Jun+Jul+Aug ... and across all customers there are probably 12 different calendars!! I use a time analysis dimension to generate YTD against my calendar YTD = aggregate(periodstodate(time.year,time.currentmemb er),[time analysis].[current]) To follow this method, I would have to provide 12 alternate hierarchies and 12 alternate YTD calculations - very messy! An extra dimension (with 12 members) to generate these YTDs might be the way to go. But it still looks like the MDX would be tricky Any other ideas would be welcome Regards Ian Bamforth |
![]() |
| Thread Tools | |
| Display Modes | |
| |