I am creating a dashboard and want to show the 3-month moving average.

I have new items being added each month and using my current approach,when a new item is added, its value is being divided by 3 but it only has a score for 1 of the 3 months in range, so it is misleading.

For example,  if the value of metric ABC is 1 for October 2014, and it was added in October 2014, it won't have values in September 2014 and August 2014 so it's 3-month moving average is is .33 if the end user selects October 2014 as the anchor month.

I've followed this approach from PowerPivotpro.com. How do I get it to not include the months where there were no values?

http://www.powerpivotpro.com/2013/07/moving-averages-sums-etc/

You can replace the "3" with a measure which calculates the number of months on the current filter context

So somthing like:

[CountMonthForMovingAvg]=CALCULATE(DISTINCTCOUNT(Calendar[MonthName]),DATESINPERIOD(Calendar[Date],LASTDATE(Calendar[Date]),-3,MONTH))