Forum Discussion

Anonymous's avatar
Anonymous
Not applicable
2 years ago
Solved

Rate of change (Trend %) per Week

Hey Everyone,

I'm trying to get a colum in a Power BI table that shows a calculated average improvement based on weekly metric results.

 

The raw data looks likes this.

 

My idea is to calculate the average metric per week using it's weight with a dax formula. Like this:

 

 

AVG Metric = 
 DIVIDE(
    SUMX(
        RAW_METRICS2,
        RAW_METRICS2[Metric] * RAW_METRICS2[Weight]
    ),
    SUM(RAW_METRICS2[Weight])
 )

 

 

 

Then, somehow calculate the "improvement" of the current week compared to the past week the trend using this formula:

 

 

week_trend =  (current week AVG metric) - (past week AVG metric)

 

 

 

And for all selected weeks between the selected dates (using slicer) get the average improvement.

 

This table roughly sumarizes the improvement week over week of selected data. 

 

This is what i want to see in the Power BI table on my report. A

 

 

 

 

  • Anonymous's avatar
    Anonymous
    2 years ago

    Hi Anonymous ,

    I create a table as you mentioned.

    Then I a measure , here is the DAX code.

    Improvement =
    VAR _currentPerson =
        SELECTEDVALUE ( RAW_METRICS2[Person] )
    VAR _currentFW =
        MAX ( 'RAW_METRICS2'[FW] )
    VAR _PreviousFW =
        CALCULATE (
            MAX ( 'RAW_METRICS2'[FW] ),
            FILTER ( ALLSELECTED ( RAW_METRICS2 ), 'RAW_METRICS2'[FW] < _currentFW )
        )
    VAR _vtable =
        SUMMARIZE (
            ALLSELECTED ( 'RAW_METRICS2' ),
            'RAW_METRICS2'[Person],
            'RAW_METRICS2'[FW],
            "_Metric", 'RAW_METRICS2'[AVG Metric]
        )
    VAR _vtable2 =
        ADDCOLUMNS (
            _vtable,
            "_LastMetric",
                MAXX (
                    FILTER ( _vtable, [Person] = _currentPerson && [FW] = _PreviousFW ),
                    [_Metric]
                )
        )
    RETURN
        IF (
            MAXX (
                FILTER ( _vtable2, [Person] = _currentPerson && [FW] = _currentFW ),
                [_LastMetric]
            )
                <> BLANK (),
            [AVG Metric]
                - MAXX (
                    FILTER ( _vtable2, [Person] = _currentPerson && [FW] = _currentFW ),
                    [_LastMetric]
                ),
            0
        )

    Finally you will get what you want.

     

     

     

    Best Regards

    Yilong Zhou

    If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

3 Replies

  • ExcelMonke's avatar
    ExcelMonke
    Impactful Individual

    If you remove the FW column it may give you your intended result

    • Anonymous's avatar
      Anonymous
      Not applicable

      Hey thank you, but what I need is the AVG improvement wow, so I can't remove the FW colum from the raw data. I do, though, not have it on the Powe BI table, but then the improvement is not calculated WOW

  • Anonymous's avatar
    Anonymous
    Not applicable

    Hi Anonymous ,

    I create a table as you mentioned.

    Then I a measure , here is the DAX code.

    Improvement =
    VAR _currentPerson =
        SELECTEDVALUE ( RAW_METRICS2[Person] )
    VAR _currentFW =
        MAX ( 'RAW_METRICS2'[FW] )
    VAR _PreviousFW =
        CALCULATE (
            MAX ( 'RAW_METRICS2'[FW] ),
            FILTER ( ALLSELECTED ( RAW_METRICS2 ), 'RAW_METRICS2'[FW] < _currentFW )
        )
    VAR _vtable =
        SUMMARIZE (
            ALLSELECTED ( 'RAW_METRICS2' ),
            'RAW_METRICS2'[Person],
            'RAW_METRICS2'[FW],
            "_Metric", 'RAW_METRICS2'[AVG Metric]
        )
    VAR _vtable2 =
        ADDCOLUMNS (
            _vtable,
            "_LastMetric",
                MAXX (
                    FILTER ( _vtable, [Person] = _currentPerson && [FW] = _PreviousFW ),
                    [_Metric]
                )
        )
    RETURN
        IF (
            MAXX (
                FILTER ( _vtable2, [Person] = _currentPerson && [FW] = _currentFW ),
                [_LastMetric]
            )
                <> BLANK (),
            [AVG Metric]
                - MAXX (
                    FILTER ( _vtable2, [Person] = _currentPerson && [FW] = _currentFW ),
                    [_LastMetric]
                ),
            0
        )

    Finally you will get what you want.

     

     

     

    Best Regards

    Yilong Zhou

    If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.