Pausal Zivference
@pausalz.bsky.social
Paul Zivich, Assistant (to the Regional) Professor
Computational epidemiologist, causal inference researcher, amateur mycologist, and open-source enthusiast.
https://github.com/pzivich
#epidemiology #statistics #python #episky #causalsky
Computational epidemiologist, causal inference researcher, amateur mycologist, and open-source enthusiast.
https://github.com/pzivich
#epidemiology #statistics #python #episky #causalsky
I should add that I am looking to give talks on this work. If you're interested (or want to know about this work but don't have time to read the paper), please reach out
I also might be able to pay my own way through some other sources, so you only need to invite me
I also might be able to pay my own way through some other sources, so you only need to invite me
I may have finally gone off the methodological deep end. A new preprint from me about some methodological ideas that have been brewing for years
I'll give a bit more background on how I got here relative to the intro, and a brief overview of the paper
arxiv.org/abs/2511.01960
I'll give a bit more background on how I got here relative to the intro, and a brief overview of the paper
arxiv.org/abs/2511.01960
Towards a Unified Framework for Statistical and Mathematical Modeling
Within the biological, physical, and social sciences, there are two broad quantitative traditions: statistical and mathematical modeling. Both traditions have the common pursuit of advancing our scien...
arxiv.org
November 13, 2025 at 4:58 PM
I should add that I am looking to give talks on this work. If you're interested (or want to know about this work but don't have time to read the paper), please reach out
I also might be able to pay my own way through some other sources, so you only need to invite me
I also might be able to pay my own way through some other sources, so you only need to invite me
Reposted by Pausal Zivference
This is one of my more out there papers. So, it is just sitting on arXiv (not submitted yet) because I am hoping to get additional feedback first
So, please consider giving it a read and letting me know your thoughts
arxiv.org/abs/2511.01960
So, please consider giving it a read and letting me know your thoughts
arxiv.org/abs/2511.01960
Towards a Unified Framework for Statistical and Mathematical Modeling
Within the biological, physical, and social sciences, there are two broad quantitative traditions: statistical and mathematical modeling. Both traditions have the common pursuit of advancing our scien...
arxiv.org
November 5, 2025 at 3:39 PM
This is one of my more out there papers. So, it is just sitting on arXiv (not submitted yet) because I am hoping to get additional feedback first
So, please consider giving it a read and letting me know your thoughts
arxiv.org/abs/2511.01960
So, please consider giving it a read and letting me know your thoughts
arxiv.org/abs/2511.01960
Reposted by Pausal Zivference
I may have finally gone off the methodological deep end. A new preprint from me about some methodological ideas that have been brewing for years
I'll give a bit more background on how I got here relative to the intro, and a brief overview of the paper
arxiv.org/abs/2511.01960
I'll give a bit more background on how I got here relative to the intro, and a brief overview of the paper
arxiv.org/abs/2511.01960
Towards a Unified Framework for Statistical and Mathematical Modeling
Within the biological, physical, and social sciences, there are two broad quantitative traditions: statistical and mathematical modeling. Both traditions have the common pursuit of advancing our scien...
arxiv.org
November 5, 2025 at 3:39 PM
I may have finally gone off the methodological deep end. A new preprint from me about some methodological ideas that have been brewing for years
I'll give a bit more background on how I got here relative to the intro, and a brief overview of the paper
arxiv.org/abs/2511.01960
I'll give a bit more background on how I got here relative to the intro, and a brief overview of the paper
arxiv.org/abs/2511.01960
I may have finally gone off the methodological deep end. A new preprint from me about some methodological ideas that have been brewing for years
I'll give a bit more background on how I got here relative to the intro, and a brief overview of the paper
arxiv.org/abs/2511.01960
I'll give a bit more background on how I got here relative to the intro, and a brief overview of the paper
arxiv.org/abs/2511.01960
Towards a Unified Framework for Statistical and Mathematical Modeling
Within the biological, physical, and social sciences, there are two broad quantitative traditions: statistical and mathematical modeling. Both traditions have the common pursuit of advancing our scien...
arxiv.org
November 5, 2025 at 3:39 PM
I may have finally gone off the methodological deep end. A new preprint from me about some methodological ideas that have been brewing for years
I'll give a bit more background on how I got here relative to the intro, and a brief overview of the paper
arxiv.org/abs/2511.01960
I'll give a bit more background on how I got here relative to the intro, and a brief overview of the paper
arxiv.org/abs/2511.01960
I've had this on my reading list for awhile, but I finally got around to it and it is excellent
It is a great introduction to bounds for missing data (that I wish I read much earlier when I was first learning about them!!)
www.sciencedirect.com/science/arti...
It is a great introduction to bounds for missing data (that I wish I read much earlier when I was first learning about them!!)
www.sciencedirect.com/science/arti...
Partial identification with missing data: concepts and findings
The traditional way to cope with missing data problems has been to combine the available data with assumptions strong enough to point-identify the pro…
www.sciencedirect.com
November 4, 2025 at 4:00 PM
I've had this on my reading list for awhile, but I finally got around to it and it is excellent
It is a great introduction to bounds for missing data (that I wish I read much earlier when I was first learning about them!!)
www.sciencedirect.com/science/arti...
It is a great introduction to bounds for missing data (that I wish I read much earlier when I was first learning about them!!)
www.sciencedirect.com/science/arti...
Reposted by Pausal Zivference
We (Audrey Renson and @pausalz.bsky.social) Just updated this paper on interference in time-varying DiD settings from a while ago: arxiv.org/abs/2405.11781. Still see papers coming out regularly about problems that I'm pretty sure this solves...
Structural Nested Mean Models Under Parallel Trends with Interference
Despite the common occurrence of interference in Difference-in-Differences (DiD) applications, standard DiD methods rely on an assumption that interference is absent, and comparatively little work has...
arxiv.org
November 3, 2025 at 3:50 PM
We (Audrey Renson and @pausalz.bsky.social) Just updated this paper on interference in time-varying DiD settings from a while ago: arxiv.org/abs/2405.11781. Still see papers coming out regularly about problems that I'm pretty sure this solves...
BetterBibTex's automatic sync feature in Zotero is such a great feature
... if it would work...
everytime I update in Zotero, BetterBib deletes all the year and journal fields for all the entries (unless I manually re-export, thus defeating the purpose)
... if it would work...
everytime I update in Zotero, BetterBib deletes all the year and journal fields for all the entries (unless I manually re-export, thus defeating the purpose)
October 31, 2025 at 3:05 PM
BetterBibTex's automatic sync feature in Zotero is such a great feature
... if it would work...
everytime I update in Zotero, BetterBib deletes all the year and journal fields for all the entries (unless I manually re-export, thus defeating the purpose)
... if it would work...
everytime I update in Zotero, BetterBib deletes all the year and journal fields for all the entries (unless I manually re-export, thus defeating the purpose)
Reposted by Pausal Zivference
If you've ever found any of my work helpful, consider donating to the Python Software Foundation
Learning Python during my PhD and translating everything between programming languages helped me build my understanding of causal inference. It is also why I know estimating equations as well as I do
Learning Python during my PhD and translating everything between programming languages helped me build my understanding of causal inference. It is also why I know estimating equations as well as I do
TLDR; The PSF has made the decision to put our community and our shared diversity, equity, and inclusion values ahead of seeking $1.5M in new revenue. Please read and share. pyfound.blogspot.com/2025/10/NSF-...
🧵
🧵
The official home of the Python Programming Language
www.python.org
October 28, 2025 at 12:18 PM
If you've ever found any of my work helpful, consider donating to the Python Software Foundation
Learning Python during my PhD and translating everything between programming languages helped me build my understanding of causal inference. It is also why I know estimating equations as well as I do
Learning Python during my PhD and translating everything between programming languages helped me build my understanding of causal inference. It is also why I know estimating equations as well as I do
If you've ever found any of my work helpful, consider donating to the Python Software Foundation
Learning Python during my PhD and translating everything between programming languages helped me build my understanding of causal inference. It is also why I know estimating equations as well as I do
Learning Python during my PhD and translating everything between programming languages helped me build my understanding of causal inference. It is also why I know estimating equations as well as I do
TLDR; The PSF has made the decision to put our community and our shared diversity, equity, and inclusion values ahead of seeking $1.5M in new revenue. Please read and share. pyfound.blogspot.com/2025/10/NSF-...
🧵
🧵
The official home of the Python Programming Language
www.python.org
October 28, 2025 at 12:18 PM
If you've ever found any of my work helpful, consider donating to the Python Software Foundation
Learning Python during my PhD and translating everything between programming languages helped me build my understanding of causal inference. It is also why I know estimating equations as well as I do
Learning Python during my PhD and translating everything between programming languages helped me build my understanding of causal inference. It is also why I know estimating equations as well as I do
Probably a good sign that all the Meta Ray-Ban AI glasses commercials just focus on their aesthetics or celebrities wearing them, and not what you could use them for
The sign of a great product!!
The sign of a great product!!
October 18, 2025 at 4:32 PM
Probably a good sign that all the Meta Ray-Ban AI glasses commercials just focus on their aesthetics or celebrities wearing them, and not what you could use them for
The sign of a great product!!
The sign of a great product!!
To close my catch-up of #MEstimatorMonday (41/52), let's talk about how I use estimating equations with synthesis models
October 17, 2025 at 2:34 PM
To close my catch-up of #MEstimatorMonday (41/52), let's talk about how I use estimating equations with synthesis models
Give it up for day 4 of my #MEstimatorMonday catch-up week (40/52). Let's go over Bespoke IVs
Like proximal causal inference, this is another way to address unmeasured confounding. For this, let's go over per-protocol effects in trials
Like proximal causal inference, this is another way to address unmeasured confounding. For this, let's go over per-protocol effects in trials
a cartoon character from spongebob squarepants is holding a crab in his hand and screaming .
ALT: a cartoon character from spongebob squarepants is holding a crab in his hand and screaming .
media.tenor.com
October 17, 2025 at 12:20 AM
Give it up for day 4 of my #MEstimatorMonday catch-up week (40/52). Let's go over Bespoke IVs
Like proximal causal inference, this is another way to address unmeasured confounding. For this, let's go over per-protocol effects in trials
Like proximal causal inference, this is another way to address unmeasured confounding. For this, let's go over per-protocol effects in trials
For the 3rd day of my #MEstimatorMonday catch-up week (39/52), let's talk proximal causal inference
We can think about proximal causal inference as an extension of the standard identification assumptions to allow for more rich data structures. Specifically, we can account for unmeasured confounding
We can think about proximal causal inference as an extension of the standard identification assumptions to allow for more rich data structures. Specifically, we can account for unmeasured confounding
October 15, 2025 at 6:38 PM
For the 3rd day of my #MEstimatorMonday catch-up week (39/52), let's talk proximal causal inference
We can think about proximal causal inference as an extension of the standard identification assumptions to allow for more rich data structures. Specifically, we can account for unmeasured confounding
We can think about proximal causal inference as an extension of the standard identification assumptions to allow for more rich data structures. Specifically, we can account for unmeasured confounding
For the 2nd day of the #MEstimatorMonday week (38/52), I'm going to workshop a notion I've been thinking about regarding the idea of stacking estimating functions
October 14, 2025 at 6:25 PM
For the 2nd day of the #MEstimatorMonday week (38/52), I'm going to workshop a notion I've been thinking about regarding the idea of stacking estimating functions
I have been bad about keeping up with #MEstimatorMonday so this will be a week of M-estimator (so 37 to 41/52)
To start, let's talk computational aspects. As you might recall from the weeks on the sandwich variance, we need to compute derivatives. One option is automatic differentiation (autodiff)
To start, let's talk computational aspects. As you might recall from the weeks on the sandwich variance, we need to compute derivatives. One option is automatic differentiation (autodiff)
October 13, 2025 at 1:21 PM
I have been bad about keeping up with #MEstimatorMonday so this will be a week of M-estimator (so 37 to 41/52)
To start, let's talk computational aspects. As you might recall from the weeks on the sandwich variance, we need to compute derivatives. One option is automatic differentiation (autodiff)
To start, let's talk computational aspects. As you might recall from the weeks on the sandwich variance, we need to compute derivatives. One option is automatic differentiation (autodiff)
That trick? Running different regression models on the same data until you get a small P-value for that super nutrient
When it comes to rice and pasta, dietitians recommend eating brown or whole grain because they're more nutritious. But you can create a super nutrient in white rice and white pasta. Here's the trick.
There's a secret superfood in white rice and pasta: Here's how to unlock it
When it comes to rice and pasta, dietitians recommend eating brown or whole grain because they're more nutritious. But you can create a super nutrient in white rice and white pasta. Here's the trick.
n.pr
October 13, 2025 at 12:12 PM
That trick? Running different regression models on the same data until you get a small P-value for that super nutrient
Reposted by Pausal Zivference
For years I had trouble following some of the discussion about confidence bands, but at ACIC this year @noahgreifer.bsky.social pointed me to a helpful paper
So you don't have to be as perplexed as I once was, we have a new pre-print introducing the key ideas
arxiv.org/abs/2510.07076
So you don't have to be as perplexed as I once was, we have a new pre-print introducing the key ideas
arxiv.org/abs/2510.07076
Confidence Regions for Multiple Outcomes, Effect Modifiers, and Other Multiple Comparisons
In epidemiology, some have argued that multiple comparison corrections are not necessary as there is rarely interest in the universal null hypothesis. From a parameter estimation perspective, epidemio...
arxiv.org
October 9, 2025 at 3:34 PM
For years I had trouble following some of the discussion about confidence bands, but at ACIC this year @noahgreifer.bsky.social pointed me to a helpful paper
So you don't have to be as perplexed as I once was, we have a new pre-print introducing the key ideas
arxiv.org/abs/2510.07076
So you don't have to be as perplexed as I once was, we have a new pre-print introducing the key ideas
arxiv.org/abs/2510.07076
For years I had trouble following some of the discussion about confidence bands, but at ACIC this year @noahgreifer.bsky.social pointed me to a helpful paper
So you don't have to be as perplexed as I once was, we have a new pre-print introducing the key ideas
arxiv.org/abs/2510.07076
So you don't have to be as perplexed as I once was, we have a new pre-print introducing the key ideas
arxiv.org/abs/2510.07076
Confidence Regions for Multiple Outcomes, Effect Modifiers, and Other Multiple Comparisons
In epidemiology, some have argued that multiple comparison corrections are not necessary as there is rarely interest in the universal null hypothesis. From a parameter estimation perspective, epidemio...
arxiv.org
October 9, 2025 at 3:34 PM
For years I had trouble following some of the discussion about confidence bands, but at ACIC this year @noahgreifer.bsky.social pointed me to a helpful paper
So you don't have to be as perplexed as I once was, we have a new pre-print introducing the key ideas
arxiv.org/abs/2510.07076
So you don't have to be as perplexed as I once was, we have a new pre-print introducing the key ideas
arxiv.org/abs/2510.07076
Aptos is such a bad font, let along one to set as the default...
September 18, 2025 at 4:52 PM
Aptos is such a bad font, let along one to set as the default...
Reposted by Pausal Zivference
Can large language models stand in for human participants?
Many social scientists seem to think so, and are already using "silicon samples" in research.
One problem: depending on the analytic decisions made, you can basically get these samples to show any effect you want.
THREAD 🧵
Many social scientists seem to think so, and are already using "silicon samples" in research.
One problem: depending on the analytic decisions made, you can basically get these samples to show any effect you want.
THREAD 🧵
The threat of analytic flexibility in using large language models to simulate human data: A call to attention
Social scientists are now using large language models to create "silicon samples" - synthetic datasets intended to stand in for human respondents, aimed at revolutionising human subjects research. How...
arxiv.org
September 18, 2025 at 7:56 AM
Can large language models stand in for human participants?
Many social scientists seem to think so, and are already using "silicon samples" in research.
One problem: depending on the analytic decisions made, you can basically get these samples to show any effect you want.
THREAD 🧵
Many social scientists seem to think so, and are already using "silicon samples" in research.
One problem: depending on the analytic decisions made, you can basically get these samples to show any effect you want.
THREAD 🧵
I'm going to do a quick #MEstimatorMonday (36/52) to show how the pooled logit EE can be used to construct inverse probability of censoring weights (IPCW)
This is based on the algorithm in my pre-print, but needs a few tweaks
arxiv.org/abs/2504.13291
This is based on the algorithm in my pre-print, but needs a few tweaks
arxiv.org/abs/2504.13291
Estimating equations for survival analysis with pooled logistic regression
Pooled logistic regression models are commonly applied in survival analysis. However, the standard implementation can be computationally demanding, which is further exacerbated when using the nonparam...
arxiv.org
September 9, 2025 at 12:29 PM
I'm going to do a quick #MEstimatorMonday (36/52) to show how the pooled logit EE can be used to construct inverse probability of censoring weights (IPCW)
This is based on the algorithm in my pre-print, but needs a few tweaks
arxiv.org/abs/2504.13291
This is based on the algorithm in my pre-print, but needs a few tweaks
arxiv.org/abs/2504.13291
A well-spent $10m per year on Bill Belichick. Glad our university leadership really knows where to invest wisely
September 2, 2025 at 12:20 PM
A well-spent $10m per year on Bill Belichick. Glad our university leadership really knows where to invest wisely
Something a bit different for #MEstimatorMonday (35/52)
I'm going to take advantage of microblogging as a format and live code a new estimating equation
I'm going to take advantage of microblogging as a format and live code a new estimating equation
September 1, 2025 at 12:47 PM
Something a bit different for #MEstimatorMonday (35/52)
I'm going to take advantage of microblogging as a format and live code a new estimating equation
I'm going to take advantage of microblogging as a format and live code a new estimating equation
I hope you're wearing your hardhat because we are building bridges this #MEstimatorMonday (34/52)
More specifically, we are going to bridge different trials together to address a single question
pubmed.ncbi.nlm.nih.gov/38110289/
More specifically, we are going to bridge different trials together to address a single question
pubmed.ncbi.nlm.nih.gov/38110289/
Fusing trial data for treatment comparisons: Single vs multi-span bridging - PubMed
While randomized controlled trials (RCTs) are critical for establishing the efficacy of new therapies, there are limitations regarding what comparisons can be made directly from trial data. RCTs are l...
pubmed.ncbi.nlm.nih.gov
August 25, 2025 at 9:30 PM
I hope you're wearing your hardhat because we are building bridges this #MEstimatorMonday (34/52)
More specifically, we are going to bridge different trials together to address a single question
pubmed.ncbi.nlm.nih.gov/38110289/
More specifically, we are going to bridge different trials together to address a single question
pubmed.ncbi.nlm.nih.gov/38110289/
Reposted by Pausal Zivference
I often find myself saying "what a time to be alive" but friends, today, defector.com/it-took-many...
It Took Many Years And Billions Of Dollars, But Microsoft Finally Invented A Calculator That Is Wrong Sometimes | Defector
It’s not AI winter just yet, though there is a distinct chill in the air. Meta is shaking up and downsizing its artificial intelligence division. A new report out of MIT finds that 95 percent of compa...
defector.com
August 20, 2025 at 4:46 PM
I often find myself saying "what a time to be alive" but friends, today, defector.com/it-took-many...