Allow plot_expected_purchases_pcc in BetaGeoModel and ModifiedBetaGeoModel

[PabloRoque]

Description

• Remove Exception in plot_expected_purchases_pcc.
• Update notebooks to now supported plots
• FIX: patch ModifiedBetaGeoNBDRV internal sim_data

Related Issue

• Closes Add plot_expected_purchases_pcc support to BG/NBD and MBG/NBD Models #1421

Checklist

• Checked that the pre-commit linting/style checks pass. Feel free to comment pre-commit.ci autofix to auto-fix.
• Included tests that prove the fix is effective or that the new feature works
• Added necessary documentation (docstrings and/or example notebooks) using numpydoc format.
• If you are a pro: each commit corresponds to a relevant logical change

---

📚 Documentation preview 📚: https://pymc-marketing--1470.org.readthedocs.build/en/1470/

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.58%. Comparing base (596e7d4) to head (d4793ab).
Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1470      +/-   ##
==========================================
- Coverage   92.58%   92.58%   -0.01%     
==========================================
  Files          52       52              
  Lines        6045     6043       -2     
==========================================
- Hits         5597     5595       -2     
  Misses        448      448              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[ColtAllen]

Why this change? (I'm not opposed, just curious.)

[ColtAllen]

Want to get your opinion on the default prior change in BetaGeoModel, but otherwise this is good to merge!

[PabloRoque]

Want to get your opinion on the default prior change in BetaGeoModel, but otherwise this is good to merge!

I) If we use :

"phi_dropout_prior": Prior("HalfFlat"),
"kappa_dropout_prior": Prior("HalfFlat"),

Then model.fit(fit_method="map") errors: Logp initial evaluation results: {'alpha': 0.0, 'r': 0.0, 'phi_dropout': 0.0, 'kappa_dropout': 0.0, 'recency_frequency': -inf} You can call model.debug() for more details.

II) If we just modify those 2 but leave "alpha_prior": Prior("HalfFlat") or "r_prior": Prior("HalfFlat") then: clv.plot_expected_purchases_ppc(model, ppc="prior") fails:

@classmethod
def rng_fn(cls, rng, size):
    raise NotImplementedError("Cannot sample from half_flat variable")

Also, we should add an Issue to experiment with the default priors.
I've noticed in #1430 that the agreement with the actuals in plot_expected_purchases_over_time is heavily dependent on the choice of priors for small datasets. We should have some sensible defaults that minimize the bias.