The ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM.
Get registeredEnhance your career with this limited time 50% discount on Fabric and Power BI exams. Ends September 15. Request your voucher.
Unable to get 3M rolling sum using datesInPeriod inspite of calender marked as date table.
I have 2 tables.
Fact table(Table)- Duplicate date keys,
Calender table - Unique keys & marked as date table
Relationship - 1(Calender) to Many(Fact table)
Below is the data .
Fact table :
Table | ||
Item number | Qty | Fact date |
121 | 45 | 1-Mar-23 |
131 | 45 | 1-Apr-23 |
141 | 30 | 1-May-23 |
151 | 40 | 1-May-23 |
161 | 50 | 1-Jul-23 |
121 | 45 | 1-Aug-23 |
Calender table:
Calender |
Date |
1-Mar-23 |
2-Mar-23 |
3-Mar-23 |
4-Mar-23 |
5-Mar-23 |
6-Mar-23 |
7-Mar-23 |
8-Mar-23 |
9-Mar-23 |
10-Mar-23 |
11-Mar-23 |
Both below dax not giving the right 2M rolling sum. It is not summing up based on dates filter.
3M DatesInPeriod =
CALCULATE(sum('Table'[Qty]),DATESINPERIOD('Calender'[Date],max('Table'[Fact date]),-2,MONTH))
3m DatesBetween =
VAR _FDate= EDATE(max('Table'[Fact date]),-1)
VAR _LDate=max('Table'[Fact date])
Return
CALCULATE(sum('Table'[Qty]),DATESBETWEEN('Calender'[Date],_FDate,_LDate))
Result :
Solved! Go to Solution.
Hi @subhashg ,
The reason it is showing incorrect figures is because the formula is referencing the fact table date column instead of the calendar table. It is recommended that we always use fields from a dimension table in our visualization instead of the fact table in order to avoid unintentional issues like this. Modifying your dax formula to reference the calendar table date instead of the fact table date produces the following output for datesbetween.
Also, for datesinperiod, the same output was produced.
I attach the pbix file for your reference:
Hi @subhashg ,
The reason it is showing incorrect figures is because the formula is referencing the fact table date column instead of the calendar table. It is recommended that we always use fields from a dimension table in our visualization instead of the fact table in order to avoid unintentional issues like this. Modifying your dax formula to reference the calendar table date instead of the fact table date produces the following output for datesbetween.
Also, for datesinperiod, the same output was produced.
I attach the pbix file for your reference:
User | Count |
---|---|
65 | |
60 | |
55 | |
54 | |
31 |
User | Count |
---|---|
180 | |
88 | |
70 | |
46 | |
46 |