What's the difference between the new version of Robyn and the original 2011 Robyn

facebookexperimental / Robyn

Robyn is an experimental, AI/ML-powered and open sourced Marketing Mix Modeling (MMM) package from Meta Marketing Science. Our mission is to democratise modeling knowledge, inspire the industry through innovation, reduce human bias in the modeling process & build a strong open source marketing science community.

https://facebookexperimental.github.io/Robyn/

MIT License

1.07k stars 322 forks source link

What's the difference between the new version of Robyn and the original 2011 Robyn #956

Open wenyanyy opened 2 months ago

wenyanyy commented 2 months ago

Describe issue

I want to ask a general questions. We have been using Robyn since 2011 when Robyn wasn't a standalone R package. I've seen a lot of new features added to Robyn, other than that, any additions or modifications regarding the algorithm? For example, is there any modification to the Budget allocation algorithm, or does it utilize SLSQP? Thanks

Looking forward to your reply. Thanks

Regina

gufengzhou commented 2 months ago

Thanks for following us since the beginning! I can't name all the changes since 2021 (not 2011). But Robyn is definitely performing better now on the objective functions, with greater finetuning levers (add_penalty_factor), improved visualisation, added confidence interval for ROAS/CPA, added immediate vs carryover split, more precise lift calibration, as well as target ROAS scenario in the budget allocator. And yes, the allocator uses SLSQP + AUGLAG.