Love these topics. Added some of these models to pymc-marketing. Very cool to see you covering them!
November 10, 2026 at 7:46 AM
Love these topics. Added some of these models to pymc-marketing. Very cool to see you covering them!
I'd just use a ODE and step through with d(temp)/d(t). I'd be surprised if it levels on the upper boundary. Check out PyMC.
November 9, 2026 at 5:30 PM
I'd just use a ODE and step through with d(temp)/d(t). I'd be surprised if it levels on the upper boundary. Check out PyMC.
New in @pymc.io 's PyMC-Marketing: IV-style designs for MMM experimentation.
BudgetModelEffect uses lift tests as instruments for spend, so the response curve is identified by designed spend changes rather than by budgets that followed demand.
BudgetModelEffect uses lift tests as instruments for spend, so the response curve is identified by designed spend changes rather than by budgets that followed demand.
November 9, 2026 at 10:23 PM
New in @pymc.io 's PyMC-Marketing: IV-style designs for MMM experimentation.
BudgetModelEffect uses lift tests as instruments for spend, so the response curve is identified by designed spend changes rather than by budgets that followed demand.
BudgetModelEffect uses lift tests as instruments for spend, so the response curve is identified by designed spend changes rather than by budgets that followed demand.
At the latest Data Science Lab Brown Bag, Dr. Juan Camilo Orduz @juanitorduz.bsky.social, @pymc-labs.bsky.social, explored how probabilistic programming and causal inference can help address these questions.
Read👇 www.hertie-school.org/en/digitalgo...
Read👇 www.hertie-school.org/en/digitalgo...
From mathematics to better decisions: researcher Juan Camilo Orduz on causal inference and AI
How do you know whether a marketing investment plan actually worked when you cannot run a controlled experiment?
www.hertie-school.org
November 11, 2026 at 8:08 AM
At the latest Data Science Lab Brown Bag, Dr. Juan Camilo Orduz @juanitorduz.bsky.social, @pymc-labs.bsky.social, explored how probabilistic programming and causal inference can help address these questions.
Read👇 www.hertie-school.org/en/digitalgo...
Read👇 www.hertie-school.org/en/digitalgo...
Join our next Data Science Brown Bag event, where Dr. Juan Camilo Orduz, Principal Data Scientist at @pymc-labs.bsky.social, will explore how probabilistic programming helps researchers and data scientists tackle real-world causal inference challenges.
👉 www.hertie-school.org/en/datascien...
👉 www.hertie-school.org/en/datascien...
November 3, 2026 at 2:19 PM
Join our next Data Science Brown Bag event, where Dr. Juan Camilo Orduz, Principal Data Scientist at @pymc-labs.bsky.social, will explore how probabilistic programming helps researchers and data scientists tackle real-world causal inference challenges.
👉 www.hertie-school.org/en/datascien...
👉 www.hertie-school.org/en/datascien...
Ian Costley (Data Scientist, Wegmans) and Luca Fiaschi (Partner, PyMC Labs) will walk through how the model was built and why Wegmans built it with a partner instead of buying one.
Free webinar, Tue Nov 10, 11 AM ET / 5 PM CET
Register: dub.sh/PwHX5pi
#Webinar #PyMC
Free webinar, Tue Nov 10, 11 AM ET / 5 PM CET
Register: dub.sh/PwHX5pi
#Webinar #PyMC
November 3, 2026 at 5:04 PM
Ian Costley (Data Scientist, Wegmans) and Luca Fiaschi (Partner, PyMC Labs) will walk through how the model was built and why Wegmans built it with a partner instead of buying one.
Free webinar, Tue Nov 10, 11 AM ET / 5 PM CET
Register: dub.sh/PwHX5pi
#Webinar #PyMC
Free webinar, Tue Nov 10, 11 AM ET / 5 PM CET
Register: dub.sh/PwHX5pi
#Webinar #PyMC
on the event stream
4. Bayesian inference: a PyMC model infers each gladiators hidden stats from public play-by-play counts, with automatic audit against the truth (coverage, skill vs. prior), and access to hidden data enforced by Postgres column permissions
5. Bayesian forecasting: probit models 4/
4. Bayesian inference: a PyMC model infers each gladiators hidden stats from public play-by-play counts, with automatic audit against the truth (coverage, skill vs. prior), and access to hidden data enforced by Postgres column permissions
5. Bayesian forecasting: probit models 4/
November 1, 2026 at 5:45 PM
on the event stream
4. Bayesian inference: a PyMC model infers each gladiators hidden stats from public play-by-play counts, with automatic audit against the truth (coverage, skill vs. prior), and access to hidden data enforced by Postgres column permissions
5. Bayesian forecasting: probit models 4/
4. Bayesian inference: a PyMC model infers each gladiators hidden stats from public play-by-play counts, with automatic audit against the truth (coverage, skill vs. prior), and access to hidden data enforced by Postgres column permissions
5. Bayesian forecasting: probit models 4/
Applied #Bayesian Modeling starts Nov 2: four live weeks of Bayesian modeling with #PyMC, taught by the people who created PyMC and Stan and wrote Think Bayes.
Instructor Evan Wimpey explains why the goal is a model you can understand and trust. dub.sh/Q3EMkkt
Instructor Evan Wimpey explains why the goal is a model you can understand and trust. dub.sh/Q3EMkkt
October 28, 2026 at 4:16 PM
Applied #Bayesian Modeling starts Nov 2: four live weeks of Bayesian modeling with #PyMC, taught by the people who created PyMC and Stan and wrote Think Bayes.
Instructor Evan Wimpey explains why the goal is a model you can understand and trust. dub.sh/Q3EMkkt
Instructor Evan Wimpey explains why the goal is a model you can understand and trust. dub.sh/Q3EMkkt
dashboards are in Latin and roman-themed, because if you're going to build a gladiator arena, you might as well commit.
If you're interested, all you need is a fork and a "docker compose up -d":
www.github.com/maggomor/gladiator-arena
#PyMC #dbt #DataScience #DataEngineering #MLOps
If you're interested, all you need is a fork and a "docker compose up -d":
www.github.com/maggomor/gladiator-arena
#PyMC #dbt #DataScience #DataEngineering #MLOps
GitHub - maggomor/gladiator-arena: A toy model to practice ETL, Warehousing, Streaming, and MLOps simulating an ancient roman gladiator arena.
A toy model to practice ETL, Warehousing, Streaming, and MLOps simulating an ancient roman gladiator arena. - maggomor/gladiator-arena
www.github.com
November 1, 2026 at 5:45 PM
dashboards are in Latin and roman-themed, because if you're going to build a gladiator arena, you might as well commit.
If you're interested, all you need is a fork and a "docker compose up -d":
www.github.com/maggomor/gladiator-arena
#PyMC #dbt #DataScience #DataEngineering #MLOps
If you're interested, all you need is a fork and a "docker compose up -d":
www.github.com/maggomor/gladiator-arena
#PyMC #dbt #DataScience #DataEngineering #MLOps
I turned the tracking data into a small number of interpretable features based on football concepts and built a simple Bayesian model that predicts play success as well as a neural net.
Data available. Code in #rstats and #pydata. Model written in Stan and PyMC.
mitchhenderson.dev/posts/2026-1...
Data available. Code in #rstats and #pydata. Model written in Stan and PyMC.
mitchhenderson.dev/posts/2026-1...
Sometimes interpretability doesn’t cost predictive power – Mitch Henderson
This was one of those times
mitchhenderson.dev
October 29, 2026 at 7:46 PM
I turned the tracking data into a small number of interpretable features based on football concepts and built a simple Bayesian model that predicts play success as well as a neural net.
Data available. Code in #rstats and #pydata. Model written in Stan and PyMC.
mitchhenderson.dev/posts/2026-1...
Data available. Code in #rstats and #pydata. Model written in Stan and PyMC.
mitchhenderson.dev/posts/2026-1...
If learning #Bayesian modeling was on your list for 2026, this is your last chance this year.
Applied Bayesian Modeling with #PyMC, our final cohort of 2026, starts Monday, Nov 2: eight live sessions.
Reserve your seat: dub.sh/QtXFqTG
Applied Bayesian Modeling with #PyMC, our final cohort of 2026, starts Monday, Nov 2: eight live sessions.
Reserve your seat: dub.sh/QtXFqTG
October 29, 2026 at 2:23 PM
If learning #Bayesian modeling was on your list for 2026, this is your last chance this year.
Applied Bayesian Modeling with #PyMC, our final cohort of 2026, starts Monday, Nov 2: eight live sessions.
Reserve your seat: dub.sh/QtXFqTG
Applied Bayesian Modeling with #PyMC, our final cohort of 2026, starts Monday, Nov 2: eight live sessions.
Reserve your seat: dub.sh/QtXFqTG
Ian Costley (Data Scientist, Wegmans) and Luca Fiaschi (Partner, PyMC Labs) will walk through how the model was built, and why Wegmans built it with a partner instead of buying one.
Free webinar, Tue Nov 10, 11 AM ET / 5 PM CET
Register: dub.sh/g2i0203
#PyMC
Free webinar, Tue Nov 10, 11 AM ET / 5 PM CET
Register: dub.sh/g2i0203
#PyMC
October 27, 2026 at 3:52 PM
Ian Costley (Data Scientist, Wegmans) and Luca Fiaschi (Partner, PyMC Labs) will walk through how the model was built, and why Wegmans built it with a partner instead of buying one.
Free webinar, Tue Nov 10, 11 AM ET / 5 PM CET
Register: dub.sh/g2i0203
#PyMC
Free webinar, Tue Nov 10, 11 AM ET / 5 PM CET
Register: dub.sh/g2i0203
#PyMC
Most rankings hand you an ordered list. Chris Fonnesbeck's new post asks how far apart two college football teams really are, and how sure we can be, using a Bayesian model in PyMC that he tests on 929 games from 2025.
Read the blog: dub.sh/i0Adfq7
Read the blog: dub.sh/i0Adfq7
October 27, 2026 at 1:56 PM
Most rankings hand you an ordered list. Chris Fonnesbeck's new post asks how far apart two college football teams really are, and how sure we can be, using a Bayesian model in PyMC that he tests on 929 games from 2025.
Read the blog: dub.sh/i0Adfq7
Read the blog: dub.sh/i0Adfq7
Learn Applied Bayesian Modeling in Python at your own pace!
#Biostatistics #Statistics #MachineLearning #ComputerScience #DataScience #Mathematics #Python #PyMC #Research #ResearchTraining #Instats
#Biostatistics #Statistics #MachineLearning #ComputerScience #DataScience #Mathematics #Python #PyMC #Research #ResearchTraining #Instats
Applied Bayesian Modeling in Python - On-Demand
Ready to bring Bayesian statistics into your research? Join this hands-on workshop for PhD students, data scientists, applied statisticians, and professional researchers. Using PyMC (version 5.25 or later), you’ll learn Bayesian fundamentals, build and fit models—from regression to hierarchical models—and interpret results using MCMC, ArviZ diagnostics, and posterior predictive checks. No prior experience with Bayesian statistics or PyMC is required; basic statistics and familiarity with Python, NumPy, pandas, and Jupyter are helpful. Attend live online via Zoom or learn on demand, with seminar materials and recordings available for 30 days, plus access to an expert-moderated Q&A forum and an Instats certificate of completion. Gain practical skills and confidence to apply Bayesian methods to your own projects.
#Biostatistics #Statistics #MachineLearning #ComputerScience #DataScience #Mathematics #Python #PyMC #Research #ResearchTraining #Instats
instats.org
October 22, 2026 at 5:21 AM
Learn Applied Bayesian Modeling in Python at your own pace!
#Biostatistics #Statistics #MachineLearning #ComputerScience #DataScience #Mathematics #Python #PyMC #Research #ResearchTraining #Instats
#Biostatistics #Statistics #MachineLearning #ComputerScience #DataScience #Mathematics #Python #PyMC #Research #ResearchTraining #Instats
Commercial site-selection tools rarely handle cannibalization well.
Learn how Wegmans and PyMC Labs solved it with an in-house Bayesian model that runs on an analyst's laptop.
Nov 10, 11 AM ET
Free webinar: dub.sh/BTCNPRT
Learn how Wegmans and PyMC Labs solved it with an in-house Bayesian model that runs on an analyst's laptop.
Nov 10, 11 AM ET
Free webinar: dub.sh/BTCNPRT
October 20, 2026 at 1:09 PM
Commercial site-selection tools rarely handle cannibalization well.
Learn how Wegmans and PyMC Labs solved it with an in-house Bayesian model that runs on an analyst's laptop.
Nov 10, 11 AM ET
Free webinar: dub.sh/BTCNPRT
Learn how Wegmans and PyMC Labs solved it with an in-house Bayesian model that runs on an analyst's laptop.
Nov 10, 11 AM ET
Free webinar: dub.sh/BTCNPRT
Linear regression assumes equal spacing. Multinomial ignores order entirely.
In his latest tutorial, @allendowney.bsky.social breaks down how to model ordered survey responses properly using PyMC: dub.sh/sS77JQf
#PyMC #DataScience
In his latest tutorial, @allendowney.bsky.social breaks down how to model ordered survey responses properly using PyMC: dub.sh/sS77JQf
#PyMC #DataScience
October 14, 2026 at 4:16 PM
Linear regression assumes equal spacing. Multinomial ignores order entirely.
In his latest tutorial, @allendowney.bsky.social breaks down how to model ordered survey responses properly using PyMC: dub.sh/sS77JQf
#PyMC #DataScience
In his latest tutorial, @allendowney.bsky.social breaks down how to model ordered survey responses properly using PyMC: dub.sh/sS77JQf
#PyMC #DataScience
PyMC Labs has a coded blog on the topic, maybe useful (warning media marketing!) www.pymc-marketing.io/en/stable/no...
MMMs and Pearl’s ladder of causal inference
In this notebook, we will explore Judea Pearl’s causal ladder and how it can be applied to solve business problems with Media Mix Models (MMMs). We will assume familiarity with the basics of MMMs i...
www.pymc-marketing.io
October 14, 2026 at 2:36 PM
PyMC Labs has a coded blog on the topic, maybe useful (warning media marketing!) www.pymc-marketing.io/en/stable/no...
Let's meet @reshamas.bsky.social ⭐️
Reshama
✨ Statistician/data science consultant and Director of Data Umbrella
✨ holds an M.S. in Statistics (Rutgers) and an M.B.A. (NYU Stern)
✨ active open-source advocate (#sklearn, #PyMC 🐍)
✨ awarded the Community Leadership Award fr…
🔗 https://bit.ly/4g7ZfAn
Reshama
✨ Statistician/data science consultant and Director of Data Umbrella
✨ holds an M.S. in Statistics (Rutgers) and an M.B.A. (NYU Stern)
✨ active open-source advocate (#sklearn, #PyMC 🐍)
✨ awarded the Community Leadership Award fr…
🔗 https://bit.ly/4g7ZfAn
Image of Reshama Shaikh
Image of Reshama Shaikh (a dark-haired person wearing an blue colored shirt). The image has a darker background with nebula and puzzle pieces matching together.
bit.ly
October 15, 2026 at 3:31 PM
Let's meet @reshamas.bsky.social ⭐️
Reshama
✨ Statistician/data science consultant and Director of Data Umbrella
✨ holds an M.S. in Statistics (Rutgers) and an M.B.A. (NYU Stern)
✨ active open-source advocate (#sklearn, #PyMC 🐍)
✨ awarded the Community Leadership Award fr…
🔗 https://bit.ly/4g7ZfAn
Reshama
✨ Statistician/data science consultant and Director of Data Umbrella
✨ holds an M.S. in Statistics (Rutgers) and an M.B.A. (NYU Stern)
✨ active open-source advocate (#sklearn, #PyMC 🐍)
✨ awarded the Community Leadership Award fr…
🔗 https://bit.ly/4g7ZfAn
Let's meet @reshamas.bsky.social ⭐️
Reshama
✨ Statistician/data science consultant and Director of Data Umbrella
✨ holds an M.S. in Statistics (Rutgers) and an M.B.A. (NYU Stern)
✨ active open-source advocate (#sklearn, #PyMC 🐍)
✨ awarded the Community Leadership Award fr…
🔗 https://bit.ly/4g7ZfAn
Reshama
✨ Statistician/data science consultant and Director of Data Umbrella
✨ holds an M.S. in Statistics (Rutgers) and an M.B.A. (NYU Stern)
✨ active open-source advocate (#sklearn, #PyMC 🐍)
✨ awarded the Community Leadership Award fr…
🔗 https://bit.ly/4g7ZfAn
Image of Reshama Shaikh
Image of Reshama Shaikh (a dark-haired person wearing an blue colored shirt). The image has a darker background with nebula and puzzle pieces matching together.
bit.ly
October 15, 2026 at 2:45 PM
Let's meet @reshamas.bsky.social ⭐️
Reshama
✨ Statistician/data science consultant and Director of Data Umbrella
✨ holds an M.S. in Statistics (Rutgers) and an M.B.A. (NYU Stern)
✨ active open-source advocate (#sklearn, #PyMC 🐍)
✨ awarded the Community Leadership Award fr…
🔗 https://bit.ly/4g7ZfAn
Reshama
✨ Statistician/data science consultant and Director of Data Umbrella
✨ holds an M.S. in Statistics (Rutgers) and an M.B.A. (NYU Stern)
✨ active open-source advocate (#sklearn, #PyMC 🐍)
✨ awarded the Community Leadership Award fr…
🔗 https://bit.ly/4g7ZfAn
Want to master probabilistic modeling for real-world problems?
@allendowney.bsky.social is co-teaching Applied Bayesian Modeling with #PyMC (Nov 2 to Dec 2). 4 weeks of live, hands-on instruction directly from the creators and maintainers.
Register here: dub.sh/mhCpq6U
#Bayesian
@allendowney.bsky.social is co-teaching Applied Bayesian Modeling with #PyMC (Nov 2 to Dec 2). 4 weeks of live, hands-on instruction directly from the creators and maintainers.
Register here: dub.sh/mhCpq6U
#Bayesian
October 14, 2026 at 4:16 PM
Want to master probabilistic modeling for real-world problems?
@allendowney.bsky.social is co-teaching Applied Bayesian Modeling with #PyMC (Nov 2 to Dec 2). 4 weeks of live, hands-on instruction directly from the creators and maintainers.
Register here: dub.sh/mhCpq6U
#Bayesian
@allendowney.bsky.social is co-teaching Applied Bayesian Modeling with #PyMC (Nov 2 to Dec 2). 4 weeks of live, hands-on instruction directly from the creators and maintainers.
Register here: dub.sh/mhCpq6U
#Bayesian
5/ PyVBMC 1.5 closes the gap to modern probabilistic programming and ML frameworks:
- You can plug in PyMC models (thanks @aloctavodia.bsky.social for discussions many years ago)
- PyVBMC also talks to ArviZ, PyTorch, Jax
@pymc.io @pymc-labs.bsky.social @arviz.bsky.social @pytorch.org
- You can plug in PyMC models (thanks @aloctavodia.bsky.social for discussions many years ago)
- PyVBMC also talks to ArviZ, PyTorch, Jax
@pymc.io @pymc-labs.bsky.social @arviz.bsky.social @pytorch.org
October 13, 2026 at 8:19 AM
5/ PyVBMC 1.5 closes the gap to modern probabilistic programming and ML frameworks:
- You can plug in PyMC models (thanks @aloctavodia.bsky.social for discussions many years ago)
- PyVBMC also talks to ArviZ, PyTorch, Jax
@pymc.io @pymc-labs.bsky.social @arviz.bsky.social @pytorch.org
- You can plug in PyMC models (thanks @aloctavodia.bsky.social for discussions many years ago)
- PyVBMC also talks to ArviZ, PyTorch, Jax
@pymc.io @pymc-labs.bsky.social @arviz.bsky.social @pytorch.org
This Thursday at Innovation Day Munich:
Christian Luhmann (PyMC Labs) & Lars-Alexander Mayer (The Marcom Engine) show how Bayesian AI + agents turn marketing analysis into real-time decisions.
Request your invite: dub.sh/D40AkeY
#DecisionIntelligence #PyMCLabs
Christian Luhmann (PyMC Labs) & Lars-Alexander Mayer (The Marcom Engine) show how Bayesian AI + agents turn marketing analysis into real-time decisions.
Request your invite: dub.sh/D40AkeY
#DecisionIntelligence #PyMCLabs
October 13, 2026 at 3:52 PM
This Thursday at Innovation Day Munich:
Christian Luhmann (PyMC Labs) & Lars-Alexander Mayer (The Marcom Engine) show how Bayesian AI + agents turn marketing analysis into real-time decisions.
Request your invite: dub.sh/D40AkeY
#DecisionIntelligence #PyMCLabs
Christian Luhmann (PyMC Labs) & Lars-Alexander Mayer (The Marcom Engine) show how Bayesian AI + agents turn marketing analysis into real-time decisions.
Request your invite: dub.sh/D40AkeY
#DecisionIntelligence #PyMCLabs
2/ What's new:
- ~4× faster (noiseless targets) / ~2× (noisy) than 1.0.4
- ~10% better metrics (single runs)
- S-VBMC built in: easily stack several runs into one more accurate posterior!
- PyMC/Arviz integration, Torch/JAX targets
- FAQ, tips, a skill for coding agents, and more
- ~4× faster (noiseless targets) / ~2× (noisy) than 1.0.4
- ~10% better metrics (single runs)
- S-VBMC built in: easily stack several runs into one more accurate posterior!
- PyMC/Arviz integration, Torch/JAX targets
- FAQ, tips, a skill for coding agents, and more
October 13, 2026 at 8:19 AM
2/ What's new:
- ~4× faster (noiseless targets) / ~2× (noisy) than 1.0.4
- ~10% better metrics (single runs)
- S-VBMC built in: easily stack several runs into one more accurate posterior!
- PyMC/Arviz integration, Torch/JAX targets
- FAQ, tips, a skill for coding agents, and more
- ~4× faster (noiseless targets) / ~2× (noisy) than 1.0.4
- ~10% better metrics (single runs)
- S-VBMC built in: easily stack several runs into one more accurate posterior!
- PyMC/Arviz integration, Torch/JAX targets
- FAQ, tips, a skill for coding agents, and more
What does a simple coin flip reveal about the way we think about probability? Spyros Georgopoulos traces the difference between Bayesian and frequentist thinking through an intuitive journey from everyday examples to PyMC.
Why You Think Like a Bayesian but Were Taught Like a Frequentist
From a chocolate bar with no price tag to a marketing mix model in PyMC, and the 200-year-old integral that stood in between.
towardsdatascience.com
October 11, 2026 at 5:04 PM
What does a simple coin flip reveal about the way we think about probability? Spyros Georgopoulos traces the difference between Bayesian and frequentist thinking through an intuitive journey from everyday examples to PyMC.
Probability can look simple until you ask what the numbers actually mean. Spyros Georgopoulos breaks down Bayesian and frequentist reasoning and connects the concepts to a real marketing mix model.
Why You Think Like a Bayesian but Were Taught Like a Frequentist
From a chocolate bar with no price tag to a marketing mix model in PyMC, and the 200-year-old integral that stood in between.
towardsdatascience.com
October 9, 2026 at 7:25 PM
Probability can look simple until you ask what the numbers actually mean. Spyros Georgopoulos breaks down Bayesian and frequentist reasoning and connects the concepts to a real marketing mix model.