How to calculate the total sum of a field in beast Mode

How can I calculate the total sum for a given row based on the condition. e.g.

for every G_Account No. I will like to calculate total Amount, for Transaction = 0 and Transaction = 1 aggregated by year:


I have used the formulala:

sum(sum(`Amount`)) over (partition by year(`Posting Date`)), however it correctly calculates for the year, but sums up both for Transaction 1 & 0.


Thank you

Answers

  • Sam1
    Sam1 Member

    I think you'd need to do that with two separate beast modes; one like this:

    case when Transaction = 1 then
    sum(sum(`Amount`)) over (partition by year(`Posting Date`))
    end
    

    and one like this:

    case when Transaction = 0 then
    sum(sum(`Amount`)) over (partition by year(`Posting Date`))
    end
    

    You may need a third for the total year, but you have the formula for that already.

  • Hi @msharma since you’re wanting to do a conditional sum you need to use a case statement inside your sum to only select the values you want

    sum(sum(case when `Transaction` = 1 then `Amount` else 0 end)) over (partition by year(`Posting Date`))

    Repeat for transaction 0

    **Was this post helpful? Click Agree or Like below**
    **Did this solve your problem? Accept it as a solution!**
  • Thank you. It works.

    I have a follow up question.

    How to add hardcoded dates that the amount is summed for a period specified, even when the chard is filtered on for different dates.

    e.g. I will like to always get the sum of transactions Type 1 between the period of Jan-01-2020 and Dec-31-2021, even when the filter on the card is set for Jan-01-2021 and Feb-28-2021


    Thank you

  • You wouldn't be able to do that on the card because the filter happens before any beast mode is applied. You'd need to aggregate your data in an ETL before hand.

    **Was this post helpful? Click Agree or Like below**
    **Did this solve your problem? Accept it as a solution!**
  • Hello, I want to do kind of the same operation but more for Type 0 to Type 70 for example.

    In the Table card, I want to see first ordered by Country and then Month. What Should I do?

    (I tried two-time order but gave me a syntax error)

    Now my table uses the following code in beast mode

    COUNT(`Mktg` in 'Yes'),1) over(PARTITION by `Country` order by MONTH(`Date`) 

    is

    Type Month Measure 1

    Type0 April 100

    Type1 April 92

    Type2 April 90

    :

    Type 70 April 60

    Type0 May 75

    Type1 May 93

    Type2 May 99

    :

    &

    The Table I want should be:

    Type Month Measure 1

    Type0 April 100

    Type0 May 92

    Type0 June 80

    :

    Type0 Dec 80

    Type 1

    :