Measure Filter not working as expected, How to involve current Value?

[FrankW]

Hi there,

 as a newbie my situation looks quiet simple, but I am fully stuck. Tthere is a table with a userdefined column containing a duration:

IntervalPlus = [MyTime]+ #duration(0,0,60,0)

With this column I want to create a kind of predictive graph with a measure:

Measure = 
CALCULATE ( 
    SUM(ExData[MyValue]),
      FILTER('ExData', ExData[MyTime] <= ExData[IntervalPlus]),
      FILTER('ExData', ExData[MyTime] >= ExData[MyTime])
)

But it looks like the SUM(ExData[MyValue]) is taken overall data, not filtered.
What am I missing?
Thanks in advance
Frank

[FrankW]

Yes of yourse.
I have a table with two columns: Time and a value witch contains an amount of events per 5 minute interval.
For each 5 Minute Interval I want to create a new Datapoint containing the next 12 Rows in Sum, to get a real one hour SUM.
Example in Excel Screenshots for 3 instead of 12 Rows:

My try was to create a new column IntervalPlus with Time plus one hour (in this example 10 Minutes) and wanted to filter values where time is lower then IntervalPlus

[FreemanZ]

hi @FrankW 

try like:

Column =

SUMX(

    FILTER(

       ExData,

       ExData[Time]>=EARLIER(ExData[Time])

   &&ExData[Time]<=[IntervalPlus]

     ),

    ExData[Value]

)

[FreemanZ · Accepted solution]

or directly like:

Column =

VAR _time = ExData[Time]

RETURN

SUMX(

    FILTER(

       ExData,

       ExData[Time]>=_time

   &&ExData[Time]<=_time+1/24/6

     ),

    ExData[Value]

)

[FrankW]

Well, this works great, thank you very much. It is not easy for me to get into this, as it is more or less a side quest. Thanks a lot

[FreemanZ]

hi @FrankW 

could you tell us more about your expectation the measure, like what do you mean by "predictive graph"?