Forum Discussion

CALvalier75's avatar
CALvalier75
New Member
8 years ago

Dynamic Multi-Row Card return 3 & 6 month projections

Hi,

I need a DAX code for a multi-row card that will return the 3 & 6 month projections from a table I have called "BWR".  Below is an example of the table BWR and in this example I would want to return November's $1,370,473 and February's $ 3,054,069.  My goal is to string them together to read above a chart "3 month risk is $1,370,473 and, 6 month risk is $3,054,069".  I've tried running 1) a FIRSTNONBLANK formula but, it won't work knowing that several of months in the Real Risk USD data have blanks, 2) the other options I thought of using would be referencing the dates but, it has to be dynamic as the the 3-6 month's will move with time or, 3) a LASTNONBLANK since the 6 month (February in this case) will always be the last nonblank since there will never be data beyond 6 months.  Any help with any of my options 1-3 or, something else would be very helpful.

 

edateReal Risk USD
2/1/2018$ 201,361
3/1/2018$ 301,321
4/1/2018$ 381,325
5/1/2018 
6/1/2018$ 901,321
7/1/2018$ 519,527
8/1/2018$ 760,286
9/1/2018 
10/1/2018 
11/1/2018$ 1,370,473
12/1/2018 
1/1/2019 
2/1/2019$ 3,054,069
3/1/2019 
4/1/2019 

 

Regards,

C

3 Replies

  • Anonymous's avatar
    Anonymous
    Not applicable
    8 years ago

    Hi CALvalier75,

     

    You can try to use below formula:

    Rolling Total=
VAR currDate =
    MAX ( Table[Date] )
VAR roll_Three =
    CALCULATE (
        SUM ( Table[Real Risk USD] ),
        FILTER (
            ALLSELECTED ( Table ),
            [Date]
                >= DATE ( YEAR ( currDate ), MONTH ( currDate ) - 3, DAY ( currDate ) )
                && [date] <= currDate
        )
    )
VAR roll_Six =
    CALCULATE (
        SUM ( Table[Real Risk USD] ),
        FILTER (
            ALLSELECTED ( Table ),
            [Date]
                >= DATE ( YEAR ( currDate ), MONTH ( currDate ) - 6, DAY ( currDate ) )
                && [date] <= currDate
        )
    )
RETURN
    "Rolling 3 month:" & roll_Three & ",Rolling 6 month:" & roll_Six

     

    Regards,

    Xiaoxin Sheng

    • CALvalier75's avatar
      CALvalier75
      New Member
      8 years ago

      Hi @Xiaoxin Sheng,

      What you provided works but, I noticed that it sums the data in the table to try and create a rolling 3 and 6 month instead of calling the actual value.  In actuality the data in the table at 3 and 6 projections are already summed which is causing a double count of the data.  What would be a replacement formula for the below line in your code that would return the actual value instead of aggregation?

      SUM ( Table[Real Risk USD] ),

       Thanks,

      CAL

      • Anonymous's avatar
        Anonymous
        Not applicable
        8 years ago

        HI CALvalier75,

         

        If your table already stored 'sum up' value, you can use max function to replace sum function to get max value.

        Rolling Total=
VAR currDate =
    MAX ( Table[Date] )
VAR roll_Three =
    CALCULATE (
        MAX( Table[Real Risk USD] ),
        FILTER (
            ALLSELECTED ( Table ),
            [Date]
                >= DATE ( YEAR ( currDate ), MONTH ( currDate ) - 3, DAY ( currDate ) )
                && [date] <= currDate
        )
    )
VAR roll_Six =
    CALCULATE (
        MAX( Table[Real Risk USD] ),
        FILTER (
            ALLSELECTED ( Table ),
            [Date]
                >= DATE ( YEAR ( currDate ), MONTH ( currDate ) - 6, DAY ( currDate ) )
                && [date] <= currDate
        )
    )
RETURN
    "Rolling 3 month:" & roll_Three & ",Rolling 6 month:" & roll_Six

         

        Regards,

        Xiaoxin Sheng