search

pymc-labs / pymc-marketing

Bayesian marketing toolbox in PyMC. Media Mix (MMM), customer lifetime value (CLV), buy-till-you-die (BTYD) models and more.
https://www.pymc-marketing.io/
Apache License 2.0
652 stars 178 forks source link

yearly seasonality of each channel #979

Open ilaria-giunti opened 2 weeks ago

ilaria-giunti commented 2 weeks ago

How can I see how the impact and coefficient of each channel change throughout the year with the TVP model?

For example, I can imagine that the parameter value is different between January and October.

wd60622 commented 2 weeks ago

Hi @ilaria-giunti,

Depends on what variables are interested in. Not all of the parameters are time-varying.

I have similar response as in #765. You should be able to take out the media_temporal_latent_multiplier variable from the posterior and multiply it the beta scaler to get an effective coefficient for each channel at each time point.

Let me know if that helps