Skip to content
Discussion options

You must be logged in to vote

Hello @gitlingjin ,

Thank you for contacting us!

The shift in observations between v1.0.7 and v1.1.1, even with the same dataset and modeling configuration, is indeed expected behavior. This is due to the new default per-channel contribution priors for non-media treatments introduced in v1.1.0.

You can replicate the previous results by setting non_media_treatments_prior_type="coefficient" in your (ModelSpec). Alternatively, you could also consider using contribution priors and customizing the prior distribution of PriorDistribution.contribution_n.

Thank you

Google Meridian Support Team

Replies: 1 comment

Comment options

You must be logged in to vote
0 replies
Answer selected by gitlingjin
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants