Colt Allen

Closing this because the closed-form PMF expressions are quite complex to code out and test, and it's uncertain how much faster this would be compared to the current sampling approach,...

I'm inclined to keep the current prior specification as-is so that model configurations aren't overly cluttered, but will leave this open for now since more testing is needed.

Given the research bot insights, I'm not sure if any work is needed on this because **Priors for `a` and `b` can already be separately specified** (and typically do better...

> What about making a helper function in order to generate the alternative priors instead of changing the defaults? Interesting; could you elaborate? This probably falls more within the scope...

This seems to happen a lot with `conda` envs: https://github.com/readthedocs/readthedocs.org/issues/6627

Hey @franzoni315, Use the `time_unit` and `time_scaler` parameters for this: ```python rfm_summary(data, customer_id_col, datetime_col, time_unit="W", # "M" & "Y" are unsupported TimeDelta units in pandas, time_scaler=4.33, # so Weeks must...

@franzoni315 sorry for not seeing this sooner; I get a lot of emails related to `pymc-marketing` notifications. > I actually discovered that this parameter is not available for the `expected_customer_lifetime_value`...

Hey @stochastic1, float64 types are required for the `xarray` operations happening under the hood. The _frequency_ feature is always a whole number, but this need not be the case for...

> Is the intended use case a weekly-level summary rather than daily? That would inform our strategy. For daily raw data spanning many years, summarizing to weekly or monthly can...

> [@awni](https://github.com/awni) I am interested, can I pick this one? Any movement on this? Otherwise I'll try my hand at vibe-coding something out. The PyMC library for Bayesian modeling supports...