Forum Discussion

MT's avatar
MT
New Member
8 years ago
Solved

Estimating monthly values from table using start date/end date

Hello everyone,

 

I have data that looks like this:

 

IdentifierStart DateEnd DateUsage Value
ABCF12341/01/201528/02/2015387
AFDV13451/01/201512/01/2015234
AFDV134513/01/201531/01/2015423

 

What I need to do is to have monthly values for each identifier, but the non-uniform start and end dates are giving me a bit of a headache. Ideal output would look like this:

 

IdentifierMonthUsage Value
ABCF1234Jan-15200
ABCF1234Feb-15187
AFDV1345Jan-15657

 

I realise this would be an estimation, but I just cant figure out a way to do this. I was thinking something along the lines of getting an avg daily value, but I have no idea how I would group those values into months.

 

Has anybody dealt with this kind of data before? Any help would be much appreciated.

 

Cheers

Need Help
  • Phil_Seamark's avatar
    Phil_Seamark
    8 years ago

    Hi MT

     

    I think this might be getting close

     

    Table = 
GROUPBY(
    FILTER(
        CROSSJOIN(
            -- Create and on the fly calendar table with a month column
            ADDCOLUMNS(
                CALENDARAUTO(),
                "Month",FORMAT([Date],"MMM-YY")
                ),
           -- Add a column to our base table that adds an average daily value
            ADDCOLUMNS(
                'Table1',
                        "Averate Usage Per Day",
                        DIVIDE(
                                'Table1'[Usage Value],
                                DATEDIFF(
                                    'Table1'[Start Date],
                                    'Table1'[End Date],DAY
                                        )
                                )
                            )
                            ),
        'Table1'[Start Date] <= [Date] && 
        'Table1'[End Date] >= [Date]
        ),
        -- Perform the actual grouping here --
        -- So Group by --
        [Identifier],
        [Month],
        -- Now sum up the Average daily usage for the above groupings --
        "Usage Value", SUMX(
                            CURRENTGROUP(),
                            [Averate Usage Per Day]
                            )
        )

    If I run this over your sample data I get the following result

     

2 Replies

  • Phil_Seamark's avatar
    Phil_Seamark
    Microsoft Employee
    8 years ago

    Hi MT

     

    I think this might be getting close

     

    Table = 
GROUPBY(
    FILTER(
        CROSSJOIN(
            -- Create and on the fly calendar table with a month column
            ADDCOLUMNS(
                CALENDARAUTO(),
                "Month",FORMAT([Date],"MMM-YY")
                ),
           -- Add a column to our base table that adds an average daily value
            ADDCOLUMNS(
                'Table1',
                        "Averate Usage Per Day",
                        DIVIDE(
                                'Table1'[Usage Value],
                                DATEDIFF(
                                    'Table1'[Start Date],
                                    'Table1'[End Date],DAY
                                        )
                                )
                            )
                            ),
        'Table1'[Start Date] <= [Date] && 
        'Table1'[End Date] >= [Date]
        ),
        -- Perform the actual grouping here --
        -- So Group by --
        [Identifier],
        [Month],
        -- Now sum up the Average daily usage for the above groupings --
        "Usage Value", SUMX(
                            CURRENTGROUP(),
                            [Averate Usage Per Day]
                            )
        )

    If I run this over your sample data I get the following result

     

  • v-ljerr-msft's avatar
    v-ljerr-msft
    Microsoft Employee
    8 years ago

    Hi MT,

     

    Have you tried the solution provided by Phil_Seamark above? Does it work in your scenario? If it works, could you accept it as solution to close this thread?

     

    If you still have any question on this issue, feel free to post here. :smileyhappy:

     

    Regards