average of a measure over a dimension along the time axis

[KV2022]

Hi, I have data like this 

I need a graph like this 

Please help.  I have to show average of TAT days by stage name along the time axis .

Days at stage in the 2nd graph = TAT days in the data table

stage in the 2nd graph = stg_nm in the table

TAT days is calculated from fact

Stg_nm, - comes from dimension(which has dim_proj_stg_key)  which is connected to dim_proj_stg_key in fact

[v-yadongf-msft · Accepted solution]

Hi @KV2022 ,

This is my test fact table:

Dim table:

Create two measures:

TAT =
CALCULATE (
    DATEDIFF ( MIN ( [Date] ), MAX ( [Date] ), DAY ),
    FILTER (
        'Fact',
        'Fact'[dim_proj_stg_key] = SELECTEDVALUE ( 'Fact'[dim_proj_stg_key] )
            && 'Fact'[YearMonth] = SELECTEDVALUE ( 'Fact'[YearMonth] )
    )
)


Average_TAT =
AVERAGEX (
    FILTER (
        ALL ( 'Fact'[YearMonth] ),
        'Fact'[YearMonth] = SELECTEDVALUE ( 'Fact'[YearMonth] )
    ),
    [TAT]
)

I think this is the result you want:

If I misunderstand your demands, please share with me some screenshots of your data after hiding sensitive information and tell me what's your expected output.

Best regards,

Yadong Fang

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

[KV2022]

Hi, thanks for the reply.  I am getting result like this. 

The TAT and Average TAT are same for every stage. 

In the data model, the stage names can be duplicate in fact meaning the stages can have iterations for every month. 

The last 2 columns  in this picture below  are minimum iteration and max iteration for a stage in a month. for eg; in Oct 2017, publisher check stage has 2 iteration. So the average should be 8 TAT days average to 4 days for 2 iterations in the stage.

But Average TAT is same as TAT . Is this correct? 

Average_TAT =

AVERAGEX (

    FILTER (

        ALL ( dim_dt[MonthYear] ),

        dim_dt[MonthYear] = SELECTEDVALUE ( dim_dt[MonthYear] )

    ),

    [TAT(Days)]

)

Thanks.