Calculating weekly running sum

Hi Wilson,

Thank you for your response.

Adding week number to the table seems like a good idea. However, unfortunately, I could achieve correct result using this approach. I have tried to:

1. filter SNAP_DT <= MAX(dim_snapshot_calendar[SNAP_DT])
2. calculate sum aggregated by snap week number

For every selected snap_dt.

I have tried some approaches, for example:

fact_revenue[NN YQ CY v2 snap_week] = 
VAR maximus = MAX ( dim_snapshot_calendar[SNAP_DT] ) 
VAR totalSplit = CALCULATE ( SUM ( fact_revenue[revenue] ), 
dim_snapshot_calendar[SNAP_DT] <= maximus ) 

return 
SUMX( 
SUMMARIZE ( fact_revenue_v2, 
dim_snapshot_calendar[SNAP_WEEK_NUM] ) 
,totalSplit) 

It returns only the value for the selected SNAP_DT and not the sum of all SNAP_DTs within the same SNAP_WEEK_NUM, where SNAP_DT <= selected SNAP_DT.

If I were to achieve simillar results using SQL, it would be:

SELECT fact_revenue[snap_week_num], SUM(fact_revenue[revenue]) 
FROM fact_revenue 
WHERE fact_revenue[snapshot_date] > '2024-02-07' 
GROUP BY fact_revenue[snap_week_num] 
; 

Could you maybe tell me, how to build an analogous DAX measure?

Thank you