A Bayesian Approach to Media Mix Modeling (Michael Johns & Zhenyu Wang)

Поделиться
HTML-код
  • Опубликовано: 9 фев 2025

Комментарии • 8

  • @goelnikhils
    @goelnikhils 2 года назад +4

    Amazing Presentation and Explanation. Congratulations

  • @vikasraina9228
    @vikasraina9228 Месяц назад

    @pymc-devs The insights you shared were really valuable and helped clarify a lot of concepts around MMM. I’ve been diving into MMM using the PyMC-Marketing MMM class, and I’m curious about setting customized priors for my model. Specifically, I want to know how I can set different priors for media channels (e.g., TV, digital ads) and control variables (e.g., seasonal effects, events like promotions).
    What’s the best way to approach this in terms of distribution choices and values for priors? Should I use a specific prior for each marketing channel and control variable, or is it better to keep them uniform? Any advice on how to tailor the priors to reflect the business context would be really helpful!

  • @wuqingli9255
    @wuqingli9255 Год назад

    truly useful and thanks for sharing this! is there any code or use case?

  • @biplobgon
    @biplobgon Год назад

    an insightful presentation, kudos to you guys!👏👏 Any suggestions on how to built this, a media mix model from scratch using lightweightMMM in python?

  • @kapil_vishwakarma
    @kapil_vishwakarma 2 года назад +1

    Link to Jin and colleagues, please.

    • @pymc-devs
      @pymc-devs  2 года назад +3

      Here is a link to the original paper, Bayesian Methods for Media Mix Modeling with Carryover and Shape Effects , research.google/pubs/pub46001/

    • @kapil_vishwakarma
      @kapil_vishwakarma 2 года назад +1

      @@pymc-devs Thank you.

    • @999nico
      @999nico Год назад

      Can you share that code at min. 19 in a GitHub repository? Thanks!