A common requirement we see on customer implementations is to be able to calculate the average of a set of values but exclude zero values. For example if we have the following set of values (10,0,0,8,0), the average excluding zeros would be (10+8)/2 = 9.
The current AVERAGE function cannot exclude zeroes and so gives an AVERAGE of 18/5=0.2 in this case.
A common workaround then is to add another line item to count non zero values (IF Data <> 0 THEN 1 ELSE 0) and use this to separately calculate the non zero average.
It would be useful to have an extra parameter to the AVERAGE function which would allow us to include or exclude zeros from the calculation and avoid this extra modelling complexity.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.