Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

MMM Predictions #1

Open
AhsanRazaGardezi opened this issue Jul 25, 2019 · 1 comment
Open

MMM Predictions #1

AhsanRazaGardezi opened this issue Jul 25, 2019 · 1 comment

Comments

@AhsanRazaGardezi
Copy link

Hello Alex,

I would like to start off by stating that I found your code very helpful, I had a few questions regarding your approach to media mix modeling.

  1. How were you able to make predictions from Stan model to calculate the ROIs
  2. You discussed a new approach to MMM on Linkedin, would you share some insight on that.
@alexhallam
Copy link
Owner

  1. ROI values are found by imputing a future promotional period with a trend, such as last year sales then calculating the difference between the Stan model prediction and the trend prediction. This difference is what media mix modelers would identify as the lift due to the media (probably debatable, but this is what is sometimes used). ROI is a function of this incremental lift and media spend.

  2. Regression with autoregressive features for the y and the media values. The Stan model takes a really long time and I felt like it did not provide that much extra useable information for real business problems. If I were to look at this again I would just do regression with some thought put into lagging the y and media variables.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@alexhallam @AhsanRazaGardezi