Power BI is turning 10! Tune in for a special live episode on July 24 with behind-the-scenes stories, product evolution highlights, and a sneak peek at what’s in store for the future.
Save the dateEnhance your career with this limited time 50% discount on Fabric and Power BI exams. Ends August 31st. Request your voucher.
@tamerj1 - sorry about the earlier post, I was a bit frazzeled when I wrote it up. Hopefully this is more clear.
I am unable to obtain the correct result from an add columns summarize pattern, when I am filtering on a dimension which does not exist within the fact table.
I am working with both forecasts and actuals, and I would like to create a version which is called "Forecast + Actuals" which is a combination of actuals and the forecast values which is after the last date of actuals.
I know you can do this via a measure like if selected value = "Forecast + Actuals" then return calculate measure, but I would prefer to do it in one measure to reduce the complexity of my measure tree.
Here is an example data set.
Date | Version | Value |
31-Oct | Forecast | 100 |
30-Sep | Forecast | 150 |
30-Sep | Actual | 120 |
31-Aug | Actual | 100 |
Here is the Version Dimension table. "Forecast + Actuals" doesn't exist in my fact table, but can be derived from the fact table through a combination of Forecast and Actuals.
Version Dimension Table |
Forecast |
Actuals |
Forecast + Actuals |
I would like a measure which would populate the following:
Version | 8/31 | 9/30 | 10/31 |
Forecast | 150 | 100 | |
Actuals | 100 | 120 | |
Forecast + Actuals | 100 | 120 | 100 |
I was trying to accomplish it like this via this measure:
DataSum =
VAR MaxActualsDate =
CALCULATE ( MAX ( FactTable[Date], Version = "Actuals" ) )
VAR SummaryData =
ADDCOLUMNS (
SUMMARIZE ( 'FactTable', 'Date'[Date], Versions[Version] ),
// this is the core table
"@SourceValue", CALCULATE ( SUM ( FactTable[Value] ) )
)
RETURN
IF (
SELECTEDVALUE ( Versions[Version] ) = "Current Forecast + Actuals",
CALCULATE ( SUMX ( SummaryData, [@SourceValue] ), [Version] = "Actuals" )
+ CALCULATE (
SUMX ( SummaryData, [@SourceValue] ),
[Version] = "Forecast",
Date[Date] > MaxActualsDate
),
SUMX ( SummaryData, [@SourceValue] )
)
I didn't find any success here - so then I decided to modify the variable table by unioning the "Forecast + Actuals" into the SummaryData variable table. When i look at this in DAX studio, the table does have "Forecast + Actuals" with values, however, I am unable to bring this into a visual.
DataSum =
var MaxActualsDate = calculate( max(FactTable[Date],Version="Actuals")
Var SummaryData = ADDCOLUMNS (
TREATAS (
UNION (
SUMMARIZE ( 'FactTable', 'Date'[Date], Versions[Version] ), // this is the core table
ADDCOLUMNS (
SUMMARIZE ( FactTable', 'Date'[Date], Versions[Version] ),
"Version", "Current Forecast + Actuals" // this is where i'm trying to add the "fake data series"
)
),
'Date'[Date],
Versions[Version]
),
"@SourceValue",
IF (
[Version] = "Current Forecast + Actuals",
CALCULATE ( SUM ( FactTable[Value] ), ALL ( Versions ), Versions[Version] = "Actuals" ),
CALCULATE ( SUM (FactTable[Value])
)
)
Return Sumx ( SummaryData, [@SourceValue])
Its something like this -
Date | Version | Currency | RWA_AIRB |
10/31 | BCL Data | USD | 100 |
10/31 | BCL Data | CAD | 100 |
10/31 | UAT | CAD | 100 |
It is still not very clear. Can we connect via teams or zoom. I'll be free after 30 min
Yeah you are right let me rework the question and post a picture of the data model
User | Count |
---|---|
22 | |
11 | |
8 | |
6 | |
6 |
User | Count |
---|---|
27 | |
13 | |
11 | |
9 | |
6 |