I would like to calcul : Stock Month -1+ Entrées Prév - Sorties Prév for each row(row = month)
The Column "Sorties Prev" is a measure in link with few measures.
I add a measure : ForecastCalculation =
VAR PrevInv = CALCULATE(SUMX('Simulation StockDP (2)'; 'Simulation StockDP (2)'[Stock]); DATEADD(DateTable[Date]; -1; MONTH))
RETURN (PrevInv + SUMX('Simulation StockDP (2)'; [Entrées Prév - Sorties Prév]))
The line for February is good, but for March the resulat is note good because the stock egals 0.
For March, i would like to calculate : Stock February + Entrées Prev - Sorties Prév (6673+2576-2693)
For April, i would like to calculate : Stock March+ Entrées Prev - Sorties Prév (6556+1730-2474)
For May, ....
Please , help me.
Thanks a lot.
DAX does not generally like recursion. See the article:
However, I wrote a blog article recently that might help with your situation:
Could you post sample data as text and expected output? Would help tremendously.
I would like this :
Stock = 7316 (january)
Stock for February = Stock January (7316)+Entrées-Sorties
Stock for March = Stock February + Entrées + Sorties
In Power BI, Sorties ia a measure link with few measures.
Hmm, well I gave this a lot of thought. Not sure I have what I need from a sample data perspective to give you a detailed solution, but I did come up with a way to emulate recursion in DAX that might work for your purposes. See here:
While recursion is elegant from a coding point of view there are usually also non-recursive solutions for most problems.
I'm not sure if I understand your issue correctly, but it looks to me like you are trying to calculate stock on hand based on looking at the Incoming (Entrées) and Outgoing (Sortes) stock. If this is the case and your period is based off a Date column in your model you could do this by calculating a lifetime to date for Incoming and Outgoing and subtract one from the other.
VAR _prevEntrées = SUMX( Filter('table1'[DateColumn], 'table1'[DateColumn] <= MAX('table1'[DateColumn]), [Entrées] )
VAR _prevSorties = SUMX( Filter('table1'[DateColumn], 'table1'[DateColumn] <= MAX('table1'[DateColumn]), [Sorties] )
RETURN _prevEntrées - _prevSorties
My measure Forecast2 is good pour January 2019 but not good results for others month.
The measure "Sorties2" included several measures and What If Paramater (Tx de controle).
The result for the Measures "Sorties2" is good .
If i add "-Sorties2" in formula "Forecast2", the result is bad.
Thanks for your help.