Jointprob community updates - Probability Basics talk, Hierarchical Models followup
Posted September 28, 2023 by daslu ‐ 2 min read
⭐ TL;DR: See you at the September standalone talk of the Jointprob – an intro to Probability basics. Also, please check out 🎥 the recording of the August talk by David MacGillivray about Bayesian Hierarchical Models. ⭐
The Jointprob community was created by Scicloj during the summer of 2022, aspiring to be a space where friends of diverse backgrounds can learn & explore Bayesian statistics and probabilistic modeling. It encourages practitioners of different proffessional and technical backgounds to meet and collaborate, mixing and combining different approaches, applications, and programming languages.
After a couple of reading journies along Bayesian Statistics books, we started a series of standalone talks.
August: Hierarchical Linear Models
On the August meetup, which was repeated twice on the 16th & 26th, David MacGillivray demonstrated Bayesian Hierarchical Models through a real-world problem.
David reproduced the results of the paper “Bayesian hierarchical model for the prediction of football results” by Gianluca Baio and Marta A. Blangiardo using PyMC 5, compared a few variations of the models, and discussed a few of the problems & solutions that arise (e.g., Neal’s funnel and non-centered reparametrization).
September: Probability Basics for Bayesian Analysis
This September, we are having a more introductory talk, reviewing basic notions of Probability Theory. In the talk, I am presenting probability spaces, random variables, distributions, and conditioning, and we see how typical Bayesian settings are phrased using these notions.
This session is repeated twice (with some changes following the discussion):
Length The sessions will be 90 minutes long. Some of us may wish to stay afterwards and chat.
Assumed Background The talk will assume some knowledge:
- basic familiarity with the R programming language
- some background in probability & statistics (say, a first intro course in a college or university)
- basic familiarity with computing discrete probabilities and expectations and with Bayes formula
- basic knowledge of calculus, specifically the notion of integral
None of these are strictly required, but this kind of background would help one in appreciating the notions presented.
Joining If you wish to be added to our calendar events, please refer to the Joining Jointprob form.