Arman Oganisian
@stablemarkets.bsky.social
Statistician | Assistant professor @ Brown University Dept of Biostatistics | Developing nonparametric Bayesian methods for causal inference.
Research site: stablemarkets.netlify.app
#statsky
Research site: stablemarkets.netlify.app
#statsky
Why I find Bayesian nonparametric causal inference compelling in one figure.
The key distinction is btwn (1) "known" vs (2) "unknown" quantities: Make inferences about (2) conditional on (1).
Want cond. avg trt effects? Condition on data, make inferences about regression lines
The key distinction is btwn (1) "known" vs (2) "unknown" quantities: Make inferences about (2) conditional on (1).
Want cond. avg trt effects? Condition on data, make inferences about regression lines
September 24, 2025 at 3:41 PM
Why I find Bayesian nonparametric causal inference compelling in one figure.
The key distinction is btwn (1) "known" vs (2) "unknown" quantities: Make inferences about (2) conditional on (1).
Want cond. avg trt effects? Condition on data, make inferences about regression lines
The key distinction is btwn (1) "known" vs (2) "unknown" quantities: Make inferences about (2) conditional on (1).
Want cond. avg trt effects? Condition on data, make inferences about regression lines
Reposted by Arman Oganisian
Arman Oganisian: Untangling Sample and Population Level Estimands in Bayesian Causal Inference https://arxiv.org/abs/2508.15016 https://arxiv.org/pdf/2508.15016 https://arxiv.org/html/2508.15016
August 22, 2025 at 6:53 AM
Arman Oganisian: Untangling Sample and Population Level Estimands in Bayesian Causal Inference https://arxiv.org/abs/2508.15016 https://arxiv.org/pdf/2508.15016 https://arxiv.org/html/2508.15016
In causal inference problems w/ sequential treatments, long stretches of time may elapse between treatment decisions
This paper, in press at Epidemiology, was really fun to write: it discusses biases that may arise & corresponding adjustment via g-methods
arxiv.org/abs/2508.21804
This paper, in press at Epidemiology, was really fun to write: it discusses biases that may arise & corresponding adjustment via g-methods
arxiv.org/abs/2508.21804
September 2, 2025 at 2:50 PM
In causal inference problems w/ sequential treatments, long stretches of time may elapse between treatment decisions
This paper, in press at Epidemiology, was really fun to write: it discusses biases that may arise & corresponding adjustment via g-methods
arxiv.org/abs/2508.21804
This paper, in press at Epidemiology, was really fun to write: it discusses biases that may arise & corresponding adjustment via g-methods
arxiv.org/abs/2508.21804
ChatGPT explains double robustness to gen z crowd.
August 25, 2025 at 12:54 PM
ChatGPT explains double robustness to gen z crowd.
I originally wrote to share with trainees but was encouraged to post it online. I address a lot of subtleties:
Why does sample-level inference need stronger assumptions? When should/n’t we impute counterfactuals? How does this differ from g-computation? Do we really need to Bayesian bootstrap?
Why does sample-level inference need stronger assumptions? When should/n’t we impute counterfactuals? How does this differ from g-computation? Do we really need to Bayesian bootstrap?
August 22, 2025 at 12:35 PM
I originally wrote to share with trainees but was encouraged to post it online. I address a lot of subtleties:
Why does sample-level inference need stronger assumptions? When should/n’t we impute counterfactuals? How does this differ from g-computation? Do we really need to Bayesian bootstrap?
Why does sample-level inference need stronger assumptions? When should/n’t we impute counterfactuals? How does this differ from g-computation? Do we really need to Bayesian bootstrap?
Another distinction between imputation of counterfactuals versus monte carlo simulations used to approximate expectations in the g-formula: In the latter, you want the variance across sims (ie approx. error) to be ≈0. In the former, variance imputation should propagate to reflect uncertainty.
August 19, 2025 at 2:05 PM
Another distinction between imputation of counterfactuals versus monte carlo simulations used to approximate expectations in the g-formula: In the latter, you want the variance across sims (ie approx. error) to be ≈0. In the former, variance imputation should propagate to reflect uncertainty.
New paper on Bayesian Diff-in-Diff methods:
www.arxiv.org/abs/2508.02970
When doing DiD, many inspect the difference in trends in the pre-period to “check” whether parallel trends (PT) holds.
But PT is fundamentally uncheckable since it must hold in the post-period as well.
What’s going on?
www.arxiv.org/abs/2508.02970
When doing DiD, many inspect the difference in trends in the pre-period to “check” whether parallel trends (PT) holds.
But PT is fundamentally uncheckable since it must hold in the post-period as well.
What’s going on?
August 10, 2025 at 6:10 PM
New paper on Bayesian Diff-in-Diff methods:
www.arxiv.org/abs/2508.02970
When doing DiD, many inspect the difference in trends in the pre-period to “check” whether parallel trends (PT) holds.
But PT is fundamentally uncheckable since it must hold in the post-period as well.
What’s going on?
www.arxiv.org/abs/2508.02970
When doing DiD, many inspect the difference in trends in the pre-period to “check” whether parallel trends (PT) holds.
But PT is fundamentally uncheckable since it must hold in the post-period as well.
What’s going on?
Reposted by Arman Oganisian
Thanks for linking! I also have a set of slides with Stan code from a recent half-day short course:
Slide deck 1 is just a primer on Bayesian inference. Slide deck 2 is on the Bayesian causal stuff.
github.com/stablemarket...
Slide deck 1 is just a primer on Bayesian inference. Slide deck 2 is on the Bayesian causal stuff.
github.com/stablemarket...
GitHub - stablemarkets/cci_institute_2025: Materials for Bayesian Causal Inference Sessions @ University of Pennsylvania's Center for Causal Inference (CCI)'s summer institute. May 29, 2025
Materials for Bayesian Causal Inference Sessions @ University of Pennsylvania's Center for Causal Inference (CCI)'s summer institute. May 29, 2025 - stablemarkets/cci_institute_2025
github.com
June 28, 2025 at 2:48 PM
Thanks for linking! I also have a set of slides with Stan code from a recent half-day short course:
Slide deck 1 is just a primer on Bayesian inference. Slide deck 2 is on the Bayesian causal stuff.
github.com/stablemarket...
Slide deck 1 is just a primer on Bayesian inference. Slide deck 2 is on the Bayesian causal stuff.
github.com/stablemarket...
I’ve seen so many instances of conflating sample and population estimands when doing Bayesian causal inference in conference talks, papers on arxiv, papers i’ve reviewed, and even published papers. People often claim to be doing one when actually doing the other.
June 27, 2025 at 7:00 PM
I’ve seen so many instances of conflating sample and population estimands when doing Bayesian causal inference in conference talks, papers on arxiv, papers i’ve reviewed, and even published papers. People often claim to be doing one when actually doing the other.
I’m teaching a 3-hour session on Bayesian causal inference at this year’s Penn Causal Inference Summer Institute, 5/27-5/30.
Virtual registration/attendance options are available.
There are sessions on a lot of other great topics - see full agenda here:
dbei.med.upenn.edu/news-events/...
#statsky
Virtual registration/attendance options are available.
There are sessions on a lot of other great topics - see full agenda here:
dbei.med.upenn.edu/news-events/...
#statsky
2025 Penn Causal Inference Summer Institute - Penn DBEI
Discover the latest news, research breakthroughs, and expert insights from Penn’s DBEI, advancing biostatistics, epidemiology, and informatics to shape population health.
dbei.med.upenn.edu
May 3, 2025 at 3:31 PM
I’m teaching a 3-hour session on Bayesian causal inference at this year’s Penn Causal Inference Summer Institute, 5/27-5/30.
Virtual registration/attendance options are available.
There are sessions on a lot of other great topics - see full agenda here:
dbei.med.upenn.edu/news-events/...
#statsky
Virtual registration/attendance options are available.
There are sessions on a lot of other great topics - see full agenda here:
dbei.med.upenn.edu/news-events/...
#statsky
Reminder to self to post my lecture notes on first-order equivalence between bayesian bootstrap SEs, frequentist bootstrap SEs, and sandwich SEs for a linear model with heteroskedastic errors
#statstab #293 The Bayesian Bootstrap
Thoughts: I need to think more on where bootstrapping makes sense in a bayesian setting. But here's a tutorial.
#stats #bayesian #bayes #bootstrap #resampling
towardsdatascience.com/the-bayesian...
Thoughts: I need to think more on where bootstrapping makes sense in a bayesian setting. But here's a tutorial.
#stats #bayesian #bayes #bootstrap #resampling
towardsdatascience.com/the-bayesian...
The Bayesian Bootstrap | Towards Data Science
A short guide to a simple and powerful extension of the bootstrap
towardsdatascience.com
April 26, 2025 at 12:39 AM
Reminder to self to post my lecture notes on first-order equivalence between bayesian bootstrap SEs, frequentist bootstrap SEs, and sandwich SEs for a linear model with heteroskedastic errors
Reposted by Arman Oganisian
Academia is cool because if you're doing it right, every paper you published in the last 3 years feels inadequate now that you understand the topic better, but it'll take 3 years to get out the version where you get it more right, and you get to do that until one day you die! Isn't that cool
April 19, 2025 at 7:11 PM
Academia is cool because if you're doing it right, every paper you published in the last 3 years feels inadequate now that you understand the topic better, but it'll take 3 years to get out the version where you get it more right, and you get to do that until one day you die! Isn't that cool
Reposted by Arman Oganisian
Congratulations to our very own Arman Oganisian, Assistant Professor of Biostatistics, for receiving the 2025 SPH Dean’s Award for Excellence in Research Collaboration! 🏆
We’re so proud to celebrate your achievement!
We’re so proud to celebrate your achievement!
April 11, 2025 at 6:38 PM
Congratulations to our very own Arman Oganisian, Assistant Professor of Biostatistics, for receiving the 2025 SPH Dean’s Award for Excellence in Research Collaboration! 🏆
We’re so proud to celebrate your achievement!
We’re so proud to celebrate your achievement!
New paper w/ Tony Linero on Bayesian causal inference:
Independent priors on propensity score & outcome models often imply a strong prior on no *measured* confounding - a prior belief that 1) we rarely hold and 2) leads to bad frequentist performance
tinyurl.com/2udmbf6a
#statsky
Independent priors on propensity score & outcome models often imply a strong prior on no *measured* confounding - a prior belief that 1) we rarely hold and 2) leads to bad frequentist performance
tinyurl.com/2udmbf6a
#statsky
Project MUSE - Priors and Propensity Scores in Bayesian Causal Inference
tinyurl.com
April 11, 2025 at 5:37 PM
New paper w/ Tony Linero on Bayesian causal inference:
Independent priors on propensity score & outcome models often imply a strong prior on no *measured* confounding - a prior belief that 1) we rarely hold and 2) leads to bad frequentist performance
tinyurl.com/2udmbf6a
#statsky
Independent priors on propensity score & outcome models often imply a strong prior on no *measured* confounding - a prior belief that 1) we rarely hold and 2) leads to bad frequentist performance
tinyurl.com/2udmbf6a
#statsky
April 3, 2025 at 2:11 PM
I’ll be at #ENAR2025 to talk about a recent paper on Bayesian causal inference with a recurrent event outcomes!
Session 50: Monday 1:45-3:30
Talk info:
www.enar.org/meetings/spr...
Full paper:
academic.oup.com/biometrics/a...
#StatsSky
Session 50: Monday 1:45-3:30
Talk info:
www.enar.org/meetings/spr...
Full paper:
academic.oup.com/biometrics/a...
#StatsSky
A Bayesian framework for causal analysis of recurrent events with timing misalignment
Abstract. Observational studies of recurrent event rates are common in biomedical statistics. Broadly, the goal is to estimate differences in event rates u
academic.oup.com
March 23, 2025 at 6:47 PM
I’ll be at #ENAR2025 to talk about a recent paper on Bayesian causal inference with a recurrent event outcomes!
Session 50: Monday 1:45-3:30
Talk info:
www.enar.org/meetings/spr...
Full paper:
academic.oup.com/biometrics/a...
#StatsSky
Session 50: Monday 1:45-3:30
Talk info:
www.enar.org/meetings/spr...
Full paper:
academic.oup.com/biometrics/a...
#StatsSky
Bayesian Causal Inference w/ survival outcomes has never been so easy!
Check out work by Biostats PhD student Han Ji now accepted at Observational Studies.
Convenient syntax, help files, custom S3 classes, & efficient MCMC via Stan in back-end
arxiv.org/pdf/2310.12358
github.com/RuBBiT-hj/ca...
Check out work by Biostats PhD student Han Ji now accepted at Observational Studies.
Convenient syntax, help files, custom S3 classes, & efficient MCMC via Stan in back-end
arxiv.org/pdf/2310.12358
github.com/RuBBiT-hj/ca...
March 10, 2025 at 2:55 PM
Bayesian Causal Inference w/ survival outcomes has never been so easy!
Check out work by Biostats PhD student Han Ji now accepted at Observational Studies.
Convenient syntax, help files, custom S3 classes, & efficient MCMC via Stan in back-end
arxiv.org/pdf/2310.12358
github.com/RuBBiT-hj/ca...
Check out work by Biostats PhD student Han Ji now accepted at Observational Studies.
Convenient syntax, help files, custom S3 classes, & efficient MCMC via Stan in back-end
arxiv.org/pdf/2310.12358
github.com/RuBBiT-hj/ca...
So many of the responses go down a Bayesian road - it’s inevitable if all models are indeed equally plausible. Reminds me of one of my favorite quotes from Radford Neal
March 9, 2025 at 7:42 PM
So many of the responses go down a Bayesian road - it’s inevitable if all models are indeed equally plausible. Reminds me of one of my favorite quotes from Radford Neal
Software update: Daniel Kowal (Cornell) was kind enough to include an implementation of our hierarchical Bayesian bootstrap in his SeBR R package (which also has other great regression tools!) - complete w/ help files and examples.
t.co/ekVZXcDqJy
t.co/LuTO9VSJfs
t.co/ekVZXcDqJy
t.co/LuTO9VSJfs
March 7, 2025 at 4:46 PM
Software update: Daniel Kowal (Cornell) was kind enough to include an implementation of our hierarchical Bayesian bootstrap in his SeBR R package (which also has other great regression tools!) - complete w/ help files and examples.
t.co/ekVZXcDqJy
t.co/LuTO9VSJfs
t.co/ekVZXcDqJy
t.co/LuTO9VSJfs
Finished drafting lecture notes on two of my favorite results in Bayesian inference:
1) The empirical Bayes derivation of the James-Stein Estimator
2) the (first order) equivalence of Bayesian bootstrap covariance, Efron’s bootstrap covariance, and the robust sandwich covariance estimators
1) The empirical Bayes derivation of the James-Stein Estimator
2) the (first order) equivalence of Bayesian bootstrap covariance, Efron’s bootstrap covariance, and the robust sandwich covariance estimators
February 9, 2025 at 5:03 PM
Finished drafting lecture notes on two of my favorite results in Bayesian inference:
1) The empirical Bayes derivation of the James-Stein Estimator
2) the (first order) equivalence of Bayesian bootstrap covariance, Efron’s bootstrap covariance, and the robust sandwich covariance estimators
1) The empirical Bayes derivation of the James-Stein Estimator
2) the (first order) equivalence of Bayesian bootstrap covariance, Efron’s bootstrap covariance, and the robust sandwich covariance estimators
Check out this new paper by Biostatistics PhD Candidate Esteban Fernández-Morales. He develops innovative Bayesian spatial shrinkage methods for causal inference with spillovers and uses it to assess the effect of Philadelphia's 2017 beverage tax.
arxiv.org/abs/2501.08231
arxiv.org/abs/2501.08231
January 21, 2025 at 5:12 PM
Check out this new paper by Biostatistics PhD Candidate Esteban Fernández-Morales. He develops innovative Bayesian spatial shrinkage methods for causal inference with spillovers and uses it to assess the effect of Philadelphia's 2017 beverage tax.
arxiv.org/abs/2501.08231
arxiv.org/abs/2501.08231