Forum Discussion

jesperxmyrup's avatar
jesperxmyrup
New Member
3 years ago
Solved

Normalize time series data to initialize at zero

Hi.

 

I have the following line chart with a cummulative PnL.

 

My issue is that I want the line to start from zero no matter what time interval I am viewing with the slicer. So substracting the first PnL observation from the PnL time series.

 

I have two columns:

- Performance[PnL] (cummulative PnL)

- Performance[ValueDateTimeUTC] (Datetime column)

 

Thanks.

 

Best, 

Jesper

 

 

 

Need Help
  • jesperxmyrup's avatar
    jesperxmyrup
    3 years ago

    I think I finally found the solution:
    The following measure did the job:

    ScaledToZero = 
VAR min_val = CALCULATE(Min(Performance[ValueDateTimeUTC]), ALLSELECTED(Performance[ValueDateTimeUTC], Performance[PnL]))
RETURN 
    Min(Performance[PnL]) - CALCULATE(
        SUM(Performance[PnL]),Performance[ValueDateTimeUTC] = min_val)

3 Replies

  • MahyarTF's avatar
    MahyarTF
    Memorable Member
    3 years ago

    Hi,

    Use the Minimum Range for Y-axis in your chart :

     

    Appreciate your Kudus.

    Please mark it as a solution if helps you.

  • jesperxmyrup's avatar
    jesperxmyrup
    New Member
    3 years ago

    Hi MahyarTF.

     

    The problem is not to change the y-axis, but to scale the time series such that it starts at y=0. I am pretty sure there is no chart setting that can do this job, but a measure has to be created.

     

    Best

    Jesper

  • jesperxmyrup's avatar
    jesperxmyrup
    New Member
    3 years ago

    I think I finally found the solution:
    The following measure did the job:

    ScaledToZero = 
VAR min_val = CALCULATE(Min(Performance[ValueDateTimeUTC]), ALLSELECTED(Performance[ValueDateTimeUTC], Performance[PnL]))
RETURN 
    Min(Performance[PnL]) - CALCULATE(
        SUM(Performance[PnL]),Performance[ValueDateTimeUTC] = min_val)