MDX - Calculated Member with EXCEPT will not roll-up at all? - mdx

I have a Calculated Member were I want to aggregate a measure by YTD. I also want the measure to roll-up on all dimensions. There is one dimension value that I dont want to be part of the aggregation. The problem is when I try to filter this out the aggregation doesnt roll-up on the other dimensions.
This works and returns the correct value for all values in the [EK DIM OBJ KONTO] dimension, and rolls-up fine on all dimensions:
Aggregate
(
YTD([DATUM D].[Period].CurrentMember)
,[Measures].[UTFALL V]
)
When I try to add an EXCEPT-statement to exclude one value in the [EK DIM OBJ KONTO] dimension I get the wrong result. The Calculated Member always returns the same answer, it does not take any notice of the dimensions I am trying to slice it with.
Aggregate
(
{YTD([DATUM D].[Period].CurrentMember)}
*
{
Except
(
[EK DIM OBJ KONTO].[KTOB2 ID].[KTOB2 ID]
,[EK DIM OBJ KONTO].[KTOB2 ID].&[-]
)
}
,[Measures].[UTFALL V]
)

This is against the AdvWorks cube but seems to be working:
WITH
MEMBER [Measures].SumYtd AS
Aggregate
(
{YTD()}
,[Measures].[Internet Sales Amount]
)
MEMBER [Measures].SumException AS
Aggregate
(
{YTD()}
*
{
Except
(
[Customer].[Customer Geography].[Country].MEMBERS
,[Customer].[Customer Geography].[Country].&[United States]
)
}
,[Measures].[Internet Sales Amount]
)
SELECT
{
[Measures].[Internet Sales Amount]
,[Measures].SumException
,[Measures].SumYtd
} ON 0
,[Date].[Calendar].[Month].&[2007]&[11].Children ON 1
FROM [Adventure Works];

Related

MDX Get dimension count for each level of hierarchy

I have an MDX query which drills into a hierarchy - I need the count of a dimension for each level. I have tried a count function but it doesnt seem to work against each level:
member [RefIDCount] as distinctCount([Incident Details].[Process Safety Classification].&[Tier 1])
select
{
[RefIDCount]
} on columns,
{
DESCENDANTS(
[Reporting Hierarchies].[Hierarchy].[Reporting Category].&[49]
-- #RepCat#
)
} on rows
FROM [Monthly Stats]
This gives me the same count for each level :
What am I doing wrong? How do I get the count to be per level?
Take a look at the example below
with member measures.t
as
[Product].[Product Categories].currentmember.level.members.count
select
{[Measures].[Internet Sales Amount],measures.t}
on 0,
descendants([Product].[Product Categories].members
)
on 1
from
[Adventure Works]
Result

MDX calculation with children and parents (contribution to total)

I got a cube with cadastral information of some neighborhoods of a city. One of the measures is 'cadastral value': the value of the land of a neighborhood.
Now, the city is divided into districts and a district, into neighborhoods (hierarchy with two levels).
The question is: I need to calculate the percentage that the value of each neighborhood represents compared with the parent (the district) and with the total (city), and also the same about the district compared with the city.
How can I do it?
I tried:
WITH MEMBER
[Measures].[Percentage] AS
([Nbh].[Nbh].[Nbh],[Measures].[Cad value].CurrentMember
/[Nbh].[Nbh].[Nbh],[Measures].[Cad value].CurrentMember.Parent)
SELECT
{[Measures].[Percentage]} ON 0,
[Nbh].[Nbh].[Nbh].AllMembers ON 1
FROM Cadastre
Where the dimensions are:
Nbh : Neighborhood
Cad value: Cadastral value
What I get is "error" in every result value, and the explanation is "Current member () : argument (0) type mismatch: expected:'level|hierarchy|dimension', got:'measure'"
What am I doing wrong?
WITH MEMBER [Measures].[Percentage] AS
([Nbh].[Nbh].[Nbh],[Measures].[Cad value].CurrentMember
/[Nbh].[Nbh].[Nbh],[Measures].[Cad value].CurrentMember.Parent)
There are two problems with this:
One, you are using a static set to compute the value of calculated member namely:
[Nbh].[Nbh].[Nbh]
This is NOT [Nbh].[Nbh].CURRENTMEMBER
Second and the bigger mistake: Using the currentmember on a measure. You can't do that. You can use Measures.CURRENTMEMBER but that's about it. It's best used on hierarchies which are used on axes.
That said, your approach is almost legit. You need to fetch the measure value for the parent.
This should work:
WITH MEMBER
[Measures].[Percentage] AS
DIVIDE
(
[Measures].[Cad value],
([Nbh].[Nbh].CURRENTMEMBER.PARENT,[Measures].[Cad value]) //Cad value for parent
)
SELECT
{[Measures].[Percentage]} ON 0,
[Nbh].[Nbh].[Nbh].AllMembers ON 1
FROM Cadastre
Try the following:
WITH MEMBER [Measures].[Parent %] AS
IIF(
[Nbh].[Nbh].CURRENTMEMBER IS [Nbh].[Nbh].[All]
,1
,[Measures].[Cad value] /
(
[Measures].[Cad value],
[Nbh].[Nbh].CurrentMember.Parent
)
), FORMAT_STRING = 'Percent'
SELECT
{[Measures].[Parent %]} ON 0,
[Nbh].[Nbh].[Nbh].AllMembers ON 1
FROM Cadastre;
Including the IIF is a defensive measure. If you switch to showing AllMembers of the complete hierarchy then the All member will be included. Here is an example using AdvWrks:
WITH
MEMBER [Measures].[Parent %] AS
IIF
(
[Product].[Subcategory].CurrentMember
IS
[Product].[Subcategory].[All Products]
,1
,
[Measures].[Internet Sales Amount]
/
(
[Measures].[Internet Sales Amount]
,[Product].[Subcategory].CurrentMember.Parent
)
)
,FORMAT_STRING = 'Percent'
SELECT
NON EMPTY
{[Measures].[Parent %]} ON 0
,NON EMPTY
[Product].[Subcategory].ALLMEMBERS ON 1
FROM [Adventure Works];
The above returns the following:
Without the IIF the expression is the following:
WITH
MEMBER [Measures].[Parent %] AS
[Measures].[Internet Sales Amount]
/
(
[Measures].[Internet Sales Amount]
,[Product].[Subcategory].CurrentMember.Parent
)
,FORMAT_STRING = 'Percent'
SELECT
NON EMPTY
{[Measures].[Parent %]} ON 0
,NON EMPTY
[Product].[Subcategory].ALLMEMBERS ON 1
FROM [Adventure Works];
We get the following because of lack of foresight:
The error is certainly because of this statement is incorrect:
WITH MEMBER
[Measures].[Percentage]
AS ([Nbh].[Nbh].[Nbh],[Measures].[Cad value].CurrentMember/[Nbh].[Nbh].[Nbh],[Measures].[Cad value].CurrentMember.Parent)
Try to make it like this
WITH MEMBER
[Measures].[Percentage]
AS ([Nbh].[Nbh].[Nbh],[Measures].[Cad value])/([Nbh].[Nbh].[Nbh],[Measures].[Cad value])

Select kpi with month on columns

I want to get two plant's oee per month in a specific year (2013), probably it's pretty trivial but I don't know how to do it:
with member [Measures].[OEE] as 'OEE'
select
{
[Equipment].[Plant Hierarchy].[Group].[DEMO_PLANT],
[Equipment].[Plant Hierarchy].[Group].[DEMO_PLANT2]
} on columns,
{
[Time].[Month]
} on rows
from ExpertPlan
where [Time].[Year].&[2013-01-01T00:00:00]
the select is ok (plants on columns and months on rows), but i'm not sure about the "with" usage to get the values be the [Measures].[OEE]
With is for custom calculated members and custom sets.
Add your measure to the slicer (WHERE clause)
SELECT
{
[Equipment].[Plant Hierarchy].[Group].[DEMO_PLANT]
,[Equipment].[Plant Hierarchy].[Group].[DEMO_PLANT2]
} ON COLUMNS
,{[Time].[Month]} ON ROWS
FROM ExpertPlan
WHERE
(
[Time].[Year].&[2013-01-01T00:00:00]
,[Measures].[OEE]
);
Here are a couple of prototyped ways of approaching this in AdvWrks cube:
SELECT
[Customer].[Country].&[United States] ON COLUMNS
,[Date].[Calendar].[Month] ON ROWS
FROM [Adventure Works]
WHERE
(
[Date].[Calendar Year].&[2007]
,[Measures].[Internet Sales Amount]
);
SELECT
[Customer].[Country].&[United States] ON COLUMNS
,Descendants
(
[Date].[Calendar].[Calendar Year].&[2007]
,[Date].[Calendar].[Month]
) ON ROWS
FROM [Adventure Works]
WHERE
[Measures].[Internet Sales Amount];
Both of the above result in this result cell set:
Adding on to the other answer, you can also employ a subselect.
SELECT
{
[Equipment].[Plant Hierarchy].[Group].[DEMO_PLANT]
,[Equipment].[Plant Hierarchy].[Group].[DEMO_PLANT2]
} ON COLUMNS
,{[Time].[Month]} ON ROWS
FROM
(
SELECT [Time].[Year].&[2013-01-01T00:00:00] ON 0
FROM [ExpertPlan]
)
WHERE
{[Measures].[OEE]}
In MDX, WHERE is the 3rd axis, although it is not too apparent. Think it to be like excel, where you have the rows, columns and values. You can put the hierarchies(attribute) on any of the axes([Measures] also behaves like any regular hierarchy). It will just change the way your final output will look like.

List all previous Years and measure till today such that all 12 Month data is available for year.Else exclude the year

We have a typical Date Dimension.
I am struggling to write an MDX to list the Dim.Year dimension member and measure contain data strictly for all month.(from Jan-Dec) otherwise exclude the year?.
This:
SELECT
[Measures].[Internet Sales Amount] ON 0
,[Date].[Calendar].[Calendar Year].MEMBERS ON 1
FROM [Adventure Works];
Returns this:
If I just now add in NON EMPTY I should get rid of 2009 and 2010:
SELECT
[Measures].[Internet Sales Amount] ON 0
,NON EMPTY
[Date].[Calendar].[Calendar Year].MEMBERS ON 1
FROM [Adventure Works];
Now returns:
The above is a very simple example.
Other scenarios might require us to get rid of nulls using other functions such as FILTER / NonEmpty / HAVING / IIF.
Edit
To just return years that have 12 months with data I've written the following (a wee bit more complicated than the above!). I will try to simplify later although you will definitely require an additional measure to achieve your goal,
WITH
MEMBER [Measures].[cntMths] AS
Count
(
NonEmpty
(
Descendants
(
[Date].[Calendar].CurrentMember
,[Date].[Calendar].[Month]
,SELF
)
,[Measures].[Internet Sales Amount]
)
)
SELECT
{
[Measures].[Internet Sales Amount]
} ON 0
,NON EMPTY
[Date].[Calendar].[Calendar Year].MEMBERS HAVING
[Measures].[cntMths] = 12 ON 1
FROM [Adventure Works];
Edit (for #sourav_agasti)
This is a solution against Adventure Works which only uses attribute hierarchies i.e. for a cube without a well formed Date dimension:
WITH
MEMBER [Measures].[cntMths] AS
Count
(
NonEmpty
(
[Date].[Calendar Year].CurrentMember
*
[Date].[Month of Year].[Month of Year].MEMBERS
,[Measures].[Internet Sales Amount]
)
)
SELECT
{} ON 0
,[Date].[Calendar Year].[Calendar Year].MEMBERS HAVING
[Measures].[cntMths] = 12 ON 1
FROM [Adventure Works];
This is the equivalent to Sourav's script using the above method:
WITH
MEMBER [Measures].[cntMths] AS
Count
(
NonEmpty
(
[Dim Year].[Year].CurrentMember
*
[Dim Month].[Month].MEMBERS
,[Measures].[Foo]
)
)
SELECT
{} ON 0
,[Dim Year].[Year].[Year].MEMBERS HAVING
[Measures].[cntMths] = 12 ON 1
FROM [Your Cube];
Further Edit
I chanced on an article last night which gives a more efficient method. Apparently Count - NonEmpty which I've done above is not tuned to work in block calculation mode whereas Sum -IIF is. So here is a quicker alternative using the later method:
WITH
MEMBER [Measures].[cntMths - S] AS
Sum
(
[Date].[Month of Year].[Month of Year].MEMBERS
,IIF
(
IsEmpty([Measures].[Internet Sales Amount])
,null
,1
)
)
SELECT
{[Measures].[cntMths - S]} ON 0
,[Date].[Calendar Year].[Calendar Year].MEMBERS HAVING
[Measures].[cntMths - S] = 12 ON 1
FROM [Adventure Works];
I took it as a challenge to write a code which gets you the result without there being any hierarchies anywhere. So here goes my attempt(after few hours of madness!).
You must be having a [Dim Month] dimension too. I am assuming the data there is stored like Dec-2014 or Dec 2014. Basically assuming that the last 4 digits give the year.
Now, one way you can figure out the years having data is to first list out all the months which don't have data and then figuring out the year for these months and then NOT displaying these years in the final output.
The below code does that for you:
WITH SET MonthsWithNoData AS
FILTER([Dim Month].[Month].CHILDREN, [Measures].[foo] = 0)
MEMBER [Measures].Year as
RIGHT([Dim Month].[Month].CURRENTMEMBER.member_value, 4)
SET SetOfYearsWithNoData AS
GENERATE(MonthsWithNoData, STRTOSET("[Dim Year].[Year].&[" + RIGHT(MonthsWithNoData.currentmember.member_value, 4) + "]"))
SET SetOfYearsWithData AS
([Dim Year].[Year].MEMBERS - SetOfYearsWithNoData)
member yearValues as
[Dim Year].[Year].currentmember.member_value
member wellFormedyearValues as
[Dim Year].[Year].currentmember.member_unique_name
SELECT {yearvalues, wellFormedyearValues} on 0,
SetOfYearsWithData on 1
from [Your cube]
I agree that the final output does not look too beautiful though. Having said that, a hierarchy would make the code much less complicated and the MDX would be much simpler. A [Calender] dimension with requisite hierarchies in place(as assumed in #whytheq's answer) would be the better approach in terms of elegance as well as performance.

MDX Scope Vs WIth member

Using the Adventure works Cube, if I run the following code (from MS example):
with
Member [Measures].[Internet Sales Amount - Range]
AS aggregate ( [Date].[Fiscal].[Date].&[20080430]:[Date].[Fiscal].[Date].&[20080502] , [Measures].[Internet Sales Amount] )
SELECT
{ [Measures].[Internet Sales Amount - Range] } ON COLUMNS,
{ [Product].[Category].Members } ON ROWS
FROM
[Adventure Works]
You get the proper results.
However, when I implement similar functionality via the SCOPE command in a calculated tab:
scope ([Product].[Category], [Measures].[Internet Sales Amount]);
this = aggregate ( [Date].[Fiscal].[Date].&[20080430]:[Date].[Fiscal].[Date].&[20080502] , [Measures].[Internet Sales Amount] );
end scope;
and then run the following mdx query:
SELECT
{ [Measures].[Internet Sales Amount] } ON COLUMNS,
{ [Product].[Category].Members } ON ROWS
FROM
[Adventure Works]
I get vastly different results. The scope appears to be ignoring the date range or ignoring the category members, i'm not sure what. I am having a similar issue on my cube when creating a new scope with a date range.
Anyone know what's going on with the scope?
Use
CREATE MEMBER CurrentCube.[Measures].[Internet Sales Amount - Range2] AS NULL;
scope ([Measures].[Internet Sales Amount - Range2]);
this = aggregate ( [Date].[Fiscal].[Date].&[20080430]:[Date].[Fiscal].[Date].&[20080502] , [Measures].[Internet Sales Amount] );
end scope;
Otherwise, there might be an infinite recursion as you reference the measure inside the scope which you use in the SCOPE itself as well.