#63 Media Mix Models & Bayes for Marketing, with Luciano Paz
Learning Bayesian Statistics
The Proety Distribution of Retention
The probability distribution o talking about here is the probability of each that happend to belong to each cluster. So you were, the distribution of retention was not unimobal. It was, you had multimodality, and you had bumps in it. And so how can you model a mess and then extract some summary informations from this? That's how what we did, was establish a diricet process.
Proudly sponsored by PyMC Labs, the Bayesian Consultancy. Book a call, or get in touch!
Inviting someone like Luciano Paz on a stats podcast is both a pleasure and a challenge — he does so many things brilliantly that you have too many questions to ask him…
In this episode, I’ve chosen — not without difficulty — to focus on the applications of Bayesian stats in the marketing industry, especially Media Mix Models. Ok, I also asked Luciano about other topics — but you know me, I like to talk…
Originally, Luciano studied physics. He then did a PhD and postdoc in neuroscience, before transitioning into industry. During his time in academia, he used stats, machine learning and data science concepts here and there, but not in a very organized way.
But at the end of his postdoc, he got into PyMC — and that’s when everything changed… He loved the community and decided to hop on board to exit academia into a better life. After leaving academia, he worked at a company that wanted to do data science but that, for privacy reasons, didn’t have a lot of data. And now, Luciano is one of the folks working full time at the PyMC Labs consultancy.
But Luciano is not only one of the cool nerds building this crazy Bayesian adventures. He also did a lot of piano and ninjutsu. Sooooo, don’t provoke him — either in the streets or at a karaoke bar…
Our theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work at https://bababrinkman.com/ !
Thank you to my Patrons for making this episode possible!
Yusuke Saito, Avi Bryant, Ero Carrera, Giuliano Cruz, Tim Gasser, James Wade, Tradd Salvo, Adam Bartonicek, William Benton, Alan O'Donnell, Mark Ormsby, James Ahloy, Robin Taylor, Thomas Wiecki, Chad Scherrer, Nathaniel Neitzke, Zwelithini Tunyiswa, Elea McDonnell Feit, Bertrand Wilden, James Thompson, Stephen Oates, Gian Luca Di Tanna, Jack Wells, Matthew Maldonado, Ian Costley, Ally Salim, Larry Gill, Joshua Duncan, Ian Moran, Paul Oreto, Colin Caprani, George Ho, Colin Carroll, Nathaniel Burbank, Michael Osthege, Rémi Louf, Clive Edelsten, Henri Wallen, Hugo Botha, Vinh Nguyen, Raul Maldonado, Marcin Elantkowski, Adam C. Smith, Will Kurt, Andrew Moskowitz, Hector Munoz, Marco Gorelli, Simon Kessell, Bradley Rode, Patrick Kelley, Rick Anderson, Casper de Bruin, Philippe Labonde, Matthew McAnear, Michael Hankin, Cameron Smith, Luis Iberico, Tomáš Frýda, Ryan Wesslen, Andreas Netti, Riley King, Aaron Jones, Yoshiyuki Hamajima, Sven De Maeyer, Michael DeCrescenzo, Fergal M, Mason Yahr, Naoya Kanai, Steven Rowland, Aubrey Clayton, Jeannine Sue, Omri Har Shemesh and Lin Yu Sha.
Visit https://www.patreon.com/learnbayesstats to unlock exclusive Bayesian swag ;)
Links from the show:
- Luciano’s website: https://lucianopaz.github.io/
- Luciano on GitHub: https://github.com/lucianopaz
- Luciano on LinkedIn: https://www.linkedin.com/in/luciano-paz-4139b5123/
- Bayesian Media Mix Modeling for Marketing Optimization: https://www.pymc-labs.io/blog-posts/bayesian-media-mix-modeling-for-marketing-optimization/
- Improving the Speed and Accuracy of Bayesian Media Mix Models: https://www.pymc-labs.io/blog-posts/reducing-customer-acquisition-costs-how-we-helped-optimizing-hellofreshs-marketing-budget/
- Speeding up HelloFresh's Bayesian AB tests by 60x: https://www.pymc-labs.io/blog-posts/bayes-is-slow-speeding-up-hellofreshs-bayesian-ab-tests-by-60x/
- PyMC Labs YouTube channel: https://www.youtube.com/c/PyMCLabs
- LBS #21 Gaussian Processes, Bayesian Neural Nets & SIR Models, with Elizaveta Semenova: https://www.learnbayesstats.com/episode/21-gaussian-processes-bayesian-neural-nets-sir-models-with-elizaveta-semenova
- Gaussian Processes approximations in PyMC: https://github.com/pymc-devs/pymc-experimental/pull/3
- Michael Betancourt, Identifying Bayesian Mixture Models: https://betanalpha.github.io/assets/case_studies/identifying_mixture_models.html
- Identifying Bayesian Mixture Models in PyMC3: https://gist.github.com/junpenglao/4d65d1a9bf80e8d371446fadda9deb7a
- Mixture Models in PyMC: https://www.pymc.io/projects/examples/en/latest/gallery.html#mixture-models
- Osvaldo Martin’s Bayesian Analysis with Python: https://www.amazon.com/dp/B07HHBCR9G
- LBS #4 Dirichlet Processes and Neurodegenerative Diseases, with Karin Knudson: https://www.learnbayesstats.com/episode/4-dirichlet-processes-and-neurodegenerative-diseases-with-karin-knudson
- Intuitive Bayes Introductory Course: https://www.intuitivebayes.com/
The AI-powered Podcast Player
