Herb Susmann
@herbps10.bsky.social
Post-doc at NYU Grossman School of Medicine (this account is solely in my personal capacity, all views are my own etc). Non-parametric statistics, causal inference, Bayesian methods. Herbsusmann.com
Check out my talk at the Online Causal Inference Seminar last week on a practical way to deal with positivity violations using bounds 👇 #causalsky
Young Researchers' Seminar: Herb Susmann and Juraj Bodík
YouTube video by Online Causal Inference Seminar
youtube.com
New joint work published with @adrianraftery.bsky.social on methods for Bayesian probabilistic projections of migration
Demography - the flagship journal of PAA@readdemography.bsky.social · 8mo agoIn “Bayesian Projection of Extant Refugee & Asylum Seeker Populations,” @herbps10.bsky.social & @adrianraftery.bsky.social propose a time-series model for projecting refugee & asylum seeker population statistics by country of origin. @nyumedpostdocs.bsky.social read.dukeupress.edu/demography/a...
The "basic" notions of semiparametric theory, from today's arxiv.org/abs/2510.18843 from Morzywolek, Gilbert, & Luedtke
Ideally letters wouldn't be required at all, but I'd settle for them only being required at a much later stage of the process after the first stage of review
Carlos Noreña@carlosfnorena.bsky.social · 10mo agoThe wild jumble of (i) prompts from institutions (with links ranging from simple to impenetrable), (ii) centralized services like Interfolio, (iii) emails for direct sending of letters, and (iv) forms that require you to type your letter in a box, is seriously crazy-making. Can't we fix this??
State of the stats job market: here's the cumulative number of stats tenure-track jobs posted on the UF Statistics Job Board so far, since August #statsky
I love living in a city full of immigrants and tons and tons of people who are not at all like me and not like each other. It makes us all better and it makes our city better. I know I’m preaching to the choir by saying this on the lib app but I sometimes just get so overwhelmed by how special it is
New preprint out on a way to handle structural and practical violations of the overlap (also known as positivity) assumption in causal inference -- as long as the outcome is bounded, we derive simple partial identification bounds on the ATE. With @alecmcclean.bsky.social and @idiaz.bsky.social
I have a new paper out on a simple way to do causal inference with left-censored outcomes. This comes up with environmental data because measurements often have a lower limit of detection -- e.g. a chemical is undetectable below a certain level www.tandfonline.com/doi/full/10....
Non-parametric treatment effect bounds for left-censored outcomes: estimating the effect of herbicide use on 2,4-D exposure
Causal inference is concerned with defining and estimating the effect of a exposure on an outcome. For example, the Average Treatment Effect (ATE), a causal inference concept, is defined as the pop...
tandfonline.com
if you are also in the niche position of needing to run a lot of simulation studies in R on slurm clusters, I have just the thing for you: github.com/herbps10/sim...
Just published: Antoine Chambaz and I did the formal work to prove you can use Super Learner (also known as model stacking) for estimating quantiles, both in i.i.d. and streaming data settings www.sciencedirect.com/science/arti...
Quantile Super Learning for independent and online settings with application to solar power forecasting
Estimating quantiles of an outcome conditional on covariates is of fundamental interest in statistics with broad application in probabilistic predicti…
sciencedirect.com
The DHS Program is officially done. As I tell my statistics students, good data is ESSENTIAL to improve the world. We can’t make things better if we don’t know the current state of things. No new DHS data collection is an incalculable loss. www.nytimes.com/2025/02/26/h...
Trump Administration Ends Global Health Research Program
The Demographic and Health Surveys were the only sources of reliable information in many countries on metrics such as mortality, nutrition and education.
nytimes.com
Heather Randell@heatherrandell.bsky.social · 2y agoThe Demographic and Health Survey (DHS) Program is the gold standard for data on health, nutrition, and socioeconomic development across the world. As with so many other critical data collection endeavors, it is currently “on pause”.
leading off my working group talk with the traveling quack to remind everyone the healthy level of skepticism they should be bringing to the table
This is a really nice and thought provoking preprint, and I think this point is largely true, and related to how strict causal inference is designed to estimate the effect of causes, but not causes of effects (or "reverse causation" as it's sometimes called www.stat.columbia.edu/~gelman/rese...)
Ben Recht@beenwrekt.bsky.social · 2y agoMy attempt to formulate a bureaucratic theory of statistics. www.argmin.net/p/bureaucrat...
This is an interesting article, and reading it made me wonder what role causal inference has in an alternative epidemiology. Causal inference gives us some nice estimators of e.g. health effects of industrial hog plants on communities, but is that really what is needed, rather than political action?
Srijon Mukhopadhyay@srijon77.bsky.social · 2y agoI love coming back to Steve Wing’s “Whose Epidemiology, whose health?” for its rich theoretical relevance and clarity! pubmed.ncbi.nlm.nih.gov/9595342/
Nice commentary summarizing some issues with non-parametric Bayes, a big one being that in practice you often end up having to place priors on very abstract objects rather than on the things you may actually have prior information about projecteuclid.org/journals/bay...
New-ish paper alert! arxiv.org/abs/2410.13522 We tackle the challenge of comparing multiple treatments when some subjects have zero prob. of receiving certain treatments. Eg, provider profiling: comparing hospitals (the “treatments”) for patient outcomes. Positivity violations are everywhere.
Fair comparisons of causal parameters with many treatments and positivity violations
Comparing outcomes across treatments is essential in medicine and public policy. To do so, researchers typically estimate a set of parameters, possibly counterfactual, with each targeting a different ...
arxiv.org
New article out -- we combined ensemble learning and adaptive conformal inference to predict emergency department arrivals in Île-de-France and provide well-calibrated prediction intervals authors.elsevier.com/a/1kEg-4xGJ-...
authors.elsevier.com
I started a list of articles, books, and tutorials (with BibTex!) for learning the semi-parametric efficiency theory relevant to causal inference: herbsusmann.com/2024/11/05/r... What have I missed? 🤔
Resources for Learning Semi-parametric Theory | Herb Susmann
herbsusmann.com
