Challenge #3: Running Averages

[spencerhong]

Phew! Hardest one yet out of 3, thanks for introducing this one early. Running averages is a great method to know. On to the next...

[RashedDS]

Understanding the problem is harder then the solution.

Here is my solution

[Maskell_Rascal]

This challenge is worded terribly, and the output data makes little to no sense. Even the example provided in the description does little to help with understanding the task at hand. This is due to how the running averages are being calculated with the Multi-Row Formula tool. Most real world scenarios would not want averages calculated this way,  

[jjrbondoc]

Spoiler

[byagelski]

my solution...

[Emil_Kos]

I calculated the running average but my output differs from the one provided in the workflow itself. 

[j_trefethen]

Had to peak at a couple different solutions here to see if I was following the instructions correctly.

[Racam1989]

Solution posted. I still don't understand why you need to 'Group By' for RM category in the 3mo trailing average but not the 6mo trailing average. If anyone knows, I would love to learn.

Spoiler

[MartinStraech]

Like many, found that that the provided solution didn't quite do the moving averages for the edge cases properly as I understood them. It's not clean, but I used if statements until there was enough months to fill in the Average() function with proper number of rows.

Spoiler

[cherrington]

My answer has a different final layout which I believe is easier to view.  Also, the official answer has incorrect average for Jan & Feb of 2009, but mine appears to be calculating correctly.  

If someone is monitoring this... LMK if my math is wrong for Jan & Feb 2009 or the official result format is preferable for a particular reason.