Forum Discussion

JarnoVisser's avatar
JarnoVisser
Helper I
7 years ago
Solved

Forecasting per quarter

Hello, I have debt collections with a start and enddate. They have different frequencies like the following table: ID Startdate Enddate Amount Frequency 1 1-1-2017 31-12-2019 10 Mon...
  • LivioLanzo's avatar
    7 years ago

    JarnoVisser

     

    You are going to need to modify your table with this DAX and create a new table:

     

    Modified Data = 
    SELECTCOLUMNS(
        GENERATE(
            Data,
            VAR Freq = Data[Frequency]
            VAR NumberOfPeriods =
                SWITCH( 
                    Freq,
                    "Monthly", DATEDIFF(  Data[Startdate], Data[Enddate], MONTH ),
                    "2 Months", INT( DATEDIFF( Data[Startdate], Data[Enddate], MONTH ) / 2 ),
                    "Quarterly", DATEDIFF( Data[Startdate], Data[Enddate], QUARTER ),
                    "Half Year", INT( DATEDIFF( Data[Startdate], Data[Enddate], MONTH ) / 6 ),
                    "Yearly", DATEDIFF( Data[Startdate], Data[Enddate], YEAR ),
                    "Once", 0
                )
                RETURN GENERATESERIES( 0, NumberOfPeriods )
        ),    
        "ID", [ID],
        "Amount", [Amount],
        "Frequency", [Frequency],
        "Date", SWITCH( 
                    [Frequency],
                    "Monthly", EDATE( [Startdate], [Value] ),
                    "2 Months", EDATE( [Startdate], [Value] * 2 ),
                    "Quarterly", EDATE( [Startdate], [Value] * 3 ),
                    "Half YEar", EDATE( [Startdate], [Value] * 6 ),
                    "Yearly", EDATE( [Startdate], [Value] * 12 ),
                    "Once", [Startdate]
                )
    )

    Then you are able to build this simple model and get the following results:

     

    Data ModelResults

     where the measure total amount is just: 

    Total Amount = SUM( 'Modified Data'[Amount] )
     
    let me know if you need to upload the pbix file