November 16, 2017

The MineThatData Forecasting Challenge: Solution #4

The fourth solution comes to us from the Daniel Wiltse. Click here to read the report.  

Here is the three month result (rolling 12 month demand):
  • Month 80 = $81.9 million.
  • Month 81 = $81.3 million.
  • Month 82 = $80.5 million.
Here is his three month forecast:
  • Month 80 = $84.5 million.
  • Month 81 = $85.6 million.
  • Month 82 = $84.5 million.
Here is his "squared error":
  • 41.3 Trillion.  Sisifo's Page Team maintains the lead and wins the competition!!

My goodness, this wasn't an easy exercise, was it?

Not at all.

Notice how hard it was to generate a forecast that showed declining sales!

Forecasting chops are sooooooo important. I'm not sure why we de-emphasize them in favor of being able to get 7 out of 10,000 to purchase from a personalized retargeting ad, but we don't always make good choices, do we?

Forecasting is where we need to spend time in 2018. We need to be able to see the future. If we can accurately see the future, we can react to the future.

This is one of the biggest reasons why I am focusing on Hillstrom's Optimizer in 2017-2018. We need to be able to see the future, and we need to be able to adjust our trajectory based on what our simulations tell us.

How about giving the four teams/individuals that participated some props for putting their guesses out there for all to see?


P.S.: On Twitter, Daniel submitted a visualization summarizing the results of a test. Here is what Daniel submitted - give it a peek and see if there's anything there you'd use to present the results of your tests, ok?




No comments:

Post a Comment

Note: Only a member of this blog may post a comment.

This Statement is Accurate, and it is Inaccurate

Here's the link for all of you who love print and think #printisback, which most obviously ... is ... not ... back. https://www.paperage...