The ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM.
Get registeredCompete to become Power BI Data Viz World Champion! First round ends August 18th. Get started.
Hi, I need help on DAX while calculating the average of Ecode. Condition would be if count of S code is matching with the distinct count of E code it will exclude while calculating average. If count of S code is not matching with the distinct count of E code then it will take distinct count of E code and based on that I need to calculate average. What should be the approach. Any help would be really appreciated.Shown here how it would be calculated
Solved! Go to Solution.
Is this what you are looking for?
Control = VAR _DisE =DISTINCTCOUNT('Table'[E Code])
VAR _CounS = COUNT('Table'[S Code])
VAR _Cont = IF(_DisE<> _CounS,_DisE)
RETURN
_Cont
Flag = IF([Control],1,0)
Final = var _t1 = SUMX(VALUES('Table'[S Code]),[Control])
VAR _t2 = SUMX(VALUES('Table'[S Code]),[Flag])
RETURN
_t1/_t2
Sample PBIX file attached
https://dropmefiles.com/EJ14P
Is this what you are looking for?
Control = VAR _DisE =DISTINCTCOUNT('Table'[E Code])
VAR _CounS = COUNT('Table'[S Code])
VAR _Cont = IF(_DisE<> _CounS,_DisE)
RETURN
_Cont
Flag = IF([Control],1,0)
Final = var _t1 = SUMX(VALUES('Table'[S Code]),[Control])
VAR _t2 = SUMX(VALUES('Table'[S Code]),[Flag])
RETURN
_t1/_t2
Sample PBIX file attached
https://dropmefiles.com/EJ14P
Here is one approach to solve the problem in DAX:
Match = COUNT(S_Code) = DISTINCTCOUNT(E_Code)
E_Code_Count = IF([Match]=TRUE(),0,DISTINCTCOUNT(E_Code))
Avg_E_Code = AVERAGEX(FILTER(ALL(Table),[E_Code_Count]>0),[E_Code])