Hi guys ,
I`m looking for a way to create a fifth quarter on my year dimension.
Actually client needs a fifth quarter to do some adjustments. For instance year 2015 2016
q1 q2 q3 q4 q1 q2 q3 q4
we have up to q1 of 2016 to do some financial adjustments for 2015. so q1 of 2016 is also the fifth quarter of 2015.
I don't know if there is a way to trick it for instance create a derivated dimension or member or whatever as q5 , I'm all hear of all kind of advise . Thx in advance
Quote from: bbtresoo on 28 Jul 2016 10:26:15 AM
Hi guys ,
I`m looking for a way to create a fifth quarter on my year dimension.
Actually client needs a fifth quarter to do some adjustments. For instance year 2015 2016
q1 q2 q3 q4 q1 q2 q3 q4
we have up to q1 of 2016 to do some financial adjustments for 2015. so q1 of 2016 is also the fifth quarter of 2015.
I don't know if there is a way to trick it for instance create a derivated dimension or member or whatever as q5 , I'm all hear of all kind of advise . Thx in advance
Hi,
There appears to be a logical flaw here. To have Q1 of 2016 also as Q5 of 2015 would mean all the measures for Q1 2016 would roll up into both years. This would give you incorrect aggregation. For this reason, Transformer (and in fact every OLAP tool I know) doesn't permit members (categories) with multiple parents.
Cheers!
MF.
Quote from: MFGF on 29 Jul 2016 04:52:07 AM
Hi,
There appears to be a logical flaw here. To have Q1 of 2016 also as Q5 of 2015 would mean all the measures for Q1 2016 would roll up into both years. This would give you incorrect aggregation. For this reason, Transformer (and in fact every OLAP tool I know) doesn't permit members (categories) with multiple parents.
Cheers!
MF.
I got ur point. Let me bring u some additionnal informations, there is two important primary keys in each fact tables publication year (2016) , date ID ( 2016-01-10) which gives you the quarter, the management rule here is : if publication year = Date ID year then ok else Q5 => adjustment for the previous year
for instance here is a case Date ID Publication year Quarter
2016-01-10 2016 Q1
2016-04-10 2016 Q2
2016-04-10 2015 Q5 of 2015 (final quarter of 2015)
so Q5 gone be populated only if Date ID year <> publication year. So there is no way we experience some aggregation issues because of thoses 2 PK, hope it is more clear.