Colt Allen

> > That's a good idea. Do you think it's worth adding to the official documentation? > > If I can find a clever and intuitive way of summarizing the...

Updated OP to add issue link for generalized seasonality.

We have a distribution block for Continuous Contractual, but no modeling options per https://github.com/pymc-labs/pymc-marketing/issues/279. I've gotten questions on LinkedIn about modeling purchases made through active memberships and/or phone app sessions,...

Is this for hierarchical priors, or something more general?

Looking back on this, I can confirm `BetaGeoModel` benefits from hyperpriors. The `a` and `b` parameters of the Beta dropout distribution seem to benefit from a hierarchical pooling approach: https://mc-stan.org/users/documentation/case-studies/pool-binary-trials.html...

Found a test dataset: https://www.brucehardie.com/datasets/hfw_trial_data.txt Results from the above linked paper can be used for testing.

> I have also found tracking plots incredibly useful for explaining to stakeholders. CLVTools has some very nice implementations: I think `plot_incremental_transactions` is the equivalent in `lifetimes`. Here's an example:...

`plot_period_transactions` is essentially plotting a posterior predictive check for customer purchase frequency:  Unless I'm mistaken, this means we can't use a `Potential` to define the likelihood. `GammaGammaModel` is fine...

Some WIP ideas for adapting `plot_period_transactions` in this issue: https://github.com/pymc-labs/pymc-marketing/issues/278

I also want to add a parameter to return the formatted data instead of the plot, in case anyone wants to build the plots in something other than `matplotlib` (this...