My table looks like this:
CPRNR | Man-years | Salary | Average pay
-----------------------------------------
XXXXX | 0,5 | 100.000 | 200.000
[Average pay] is a measure based on [Salary]/[Man-years]. What I want is to get standard deviation for [Average pay].
I've tried the following, but it results in an error.
=STDEVX.P(DataCSV , [Average pay])
The error message is:
MdxScript(Model) (6, 34) Calculations error in measure 'DataCSV'[Average pay]: A division by zero has been encountered.
There are 0 values in [Man-years] and for the table I have added a value filter on [Cprnr] ([Man-years] >= 0,825) and I want the standard deviation to only use the values that are left.
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…