Hanna van der Vlis - Clusterf*ck: A Practical Guide to Bayesian Hierarchical Modeling in PyMC3

Hanna van der Vlis Presents:
Clusterf*ck: A Practical Guide to Bayesian Hierarchical Modeling in PyMC3
At Apollo Agriculture, a Kenya based agro-tech startup, one of the challenging problems we face is to predict yields of Kenyan maize farmers. Like almost all data-sets, this data-set has a hierarchical structure: farmers within the same region aren’t independent. By ignoring this fact, a model could predict yields entirely from the region of the farmer, but fails to find any other meaningful insights, and we may not even realize. However, if we “overcorrected,” treating each region as completely separate, each individual analysis could be underpowered. Enter the hero of our story: Bayesian hierarchical modeling. Using a practical example in Pymc3, we’ll follow this hero as they identify and overcome clustered data-sets.
Slides: pydata.org/london2022/wp-cont...
www.pydata.org
PyData is an educational program of NumFOCUS, a 501(c)3 non-profit organization in the United States. PyData provides a forum for the international community of users and developers of data analysis tools to share ideas and learn from each other. The global PyData network promotes discussion of best practices, new approaches, and emerging technologies for data management, processing, analytics, and visualization. PyData communities approach data science using many languages, including (but not limited to) Python, Julia, and R.
PyData conferences aim to be accessible and community-driven, with novice to advanced level presentations. PyData tutorials and talks bring attendees the latest project features along with cutting-edge use cases.
00:00 Welcome!
Want to help add timestamps to our UA-cam videos to help with discoverability? Find out more here: github.com/numfocus/UA-camVi...

[@VALedu11]

Hanna van der Vlis your way of creating the slides and its content is super. no formality...simply form the heart .. WTF is conjugate.. I went through that years ago.. didnt think of puttin it on my slides though.. fantastic work....wish you moreeffective PRIORS

[@mikiallen7733]

why didn't you use actually "HMMs" given that you have more experience with frequentist's approach to statistics ? in fact it allows conditioning your models on the most important drivers of a certain problem if you are interested more in to having posterior probabilities given the group characteristics !

[@devilasks]

Very confusing, better start with the sources mentioned in the slides

[@in5minutes556]

Absolutely disgusting that this sort of unprofessionalism is allowed and encouraged at pyData. What is even the point of including profanity in the title and then saying “oh, I can’t pronounce the title”

[@matthewdrury6443]

It made me laugh. I think that's the point. Humor.

[@eddwebster7163]

Grow up

[@matthewdrury6443]

I'm good. I'd rather enjoy things than clutch my pearls.

[@Eizengoldt]

Lol seethe more