Sign in

damiano cerasuolo

@damiacer.bsky.social
69 followers 61 following 70 posts

medicinæ doctor in public health/biostatistics. statistics, #rstats, #clinicalresearch, photography and random stuff. caen, paris and rome mastodon: scholar.social/@damiacer github/io: damiacer.github.io

PostsRepliesMedia
Reposted by damiano cerasuolo
ArXiv Paperboy (Stat.ME+Econ.EM) @paperposterbot.bsky.social · 21/07/2026
arXiv📈🤖 Evaluation of Missing Data Analytical Techniques in Longitudinal Research: Traditional and Machine Learning Approaches By
arXiv:2406.13814v1 Announce Type: cross 
Abstract: Missing Not at Random (MNAR) and nonnormal data are challenging to handle. Traditional missing data analytical techniques such as full information maximum likelihood estimation (FIML) may fail with nonnormal data as they are built on normal distribution assumptions. Two-Stage Robust Estimation (TSRE) does manage nonnormal data, but both FIML and TSRE are less explored in longitudinal studies under MNAR conditions with nonnormal distributions. Unlike traditional statistical approaches, machine learning approaches do not require distributional assumptions about the data. More importantly, they have shown promise for MNAR data; however, their application in longitudinal studies, addressing both Missing at Random (MAR) and MNAR scenarios, is also underexplored. This study utilizes Monte Carlo simulations to assess and compare the effectiveness of six analytical techniques for missing data within the growth curve modeling framework. These techniques include traditional approaches like FIML and TSRE, machine learning approaches by single imputation (K-Nearest Neighbors and missForest), and machine learning approaches by multiple imputation (micecart and miceForest). We investigate the influence of sample size, missing data rate, missing data mechanism, and data distribution on the accuracy and efficiency of model estimation. Our findings indicate that FIML is most effective for MNAR data among the tested approaches. TSRE excels in handling MAR data, while missForest is only advantageous in limited conditions with a combination of very skewed distributions, very large sample sizes (e.g., n larger than 1000), and low missing data rates.
021
Reposted by damiano cerasuolo
Jacobin @jacobinmag.bsky.social · 21/07/2026
Twenty-five years ago, 300,000 people marched in Genoa against the G8 summit. The Italian state responded with tear gas, rubber bullets, and improvised torture chambers, but history has vindicated the critique of capitalist globalization by the protesters.
jacobin.com
The Genoa Protesters Were Right About Global Capitalism
Twenty-five years ago, 300,000 people marched in Genoa against the G8 summit. The Italian state responded with tear gas, rubber bullets, and improvised torture chambers, but history has vindicated the critique of capitalist globalization by the protesters.
1218
Reposted by damiano cerasuolo
Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 09/07/2026
The correlation coefficient is not a proper scoring rule. An agent who ranks models by correlation can be beaten by an infinite class of alternative rules. Here's a 2007 article on proper scoring rules: sites.stat.washington.edu/raftery/Rese...
310724
damiano cerasuolo @damiacer.bsky.social · 26/06/2026
Je n'ai jamais autant visité le site de Météo France comme ces deux derniers jours.
100
Reposted by damiano cerasuolo
Isabella Velásquez @ivelasq3.bsky.social · 30/04/2026
@nrennie.bsky.social just shared a blog post on some great ggplot2 functions! bsky.app/profile/nren...
062
Reposted by damiano cerasuolo
Jelena Pantel @jhpantel.bsky.social · 30/04/2026
Now very motivated to derive the Norma distribution #rstats #stan
Error message in Stan, "No distribution 'norma' was found."
062
Reposted by damiano cerasuolo
Michael Friendly @datavisfriendly.bsky.social · 04/04/2026
VIF is often very subtle. The important info lies in the space of the smallest PCA dimensions. See my {VisCollin} #rstats pkg, friendly.github.io/VisCollin/
friendly.github.io
Visualizing Collinearity Diagnostics
Provides methods to calculate diagnostics for multicollinearity among predictors in a linear or generalized linear model. It also provides methods to visualize those diagnostics following Friendly & K...
091
Reposted by damiano cerasuolo
Andrew Gelman et al. @statmodeling.bsky.social · 25/03/2026
Notation corner: When you have several different expressions that are mathematically equivalent, you don’t have to choose just one! statmodeling.stat.columbia.edu/2026/03/25/n...
statmodeling.stat.columbia.edu
Notation corner: When you have several different expressions that are mathematically equivalent, you don’t have to choose just one! | Statistical Modeling, Causal Inference, and Social Science
042
damiano cerasuolo @damiacer.bsky.social · 13/03/2026
That’s a great news for the R community! Congratulations @heathrturnr.fosstodon.org.ap.brid.gy!
000
Reposted by damiano cerasuolo
Michael Friendly @datavisfriendly.bsky.social · 13/03/2026
#rstats #dataviz #psy6136 Lecture 9 for my course in Categorical Data Analysis-- Count Data Models 📋 Materials: friendly.github.io/psy6136/#GLM... 🎞️ Slides: friendly.github.io/psy6136/lect...
1286
Reposted by damiano cerasuolo
Olivier Gimenez 🦦 @oaggimenez.bsky.social · 01/02/2026
📘 New book out soon ! I’m excited to share that 𝐁𝐚𝐲𝐞𝐬𝐢𝐚𝐧 𝐀𝐧𝐚𝐥𝐲𝐬𝐢𝐬 𝐨𝐟 𝐂𝐚𝐩𝐭𝐮𝐫𝐞-𝐑𝐞𝐜𝐚𝐩𝐭𝐮𝐫𝐞 𝐃𝐚𝐭𝐚 𝐰𝐢𝐭𝐡 𝐇𝐢𝐝𝐝𝐞𝐧 𝐌𝐚𝐫𝐤𝐨𝐯 𝐌𝐨𝐝𝐞𝐥𝐬: 𝐓𝐡𝐞𝐨𝐫𝐲 𝐚𝐧𝐝 𝐂𝐚𝐬𝐞 𝐒𝐭𝐮𝐝𝐢𝐞𝐬 𝐢𝐧 𝐑 𝐚𝐧𝐝 𝐍𝐈𝐌𝐁𝐋𝐄 is being published by Chapman & Hall / CRC Press Hope it’s useful to students, researchers, and practitioners #StatisticalEcology #NIMBLE
Book cover for https://oliviergimenez.github.io/banana-book/
28727
Reposted by damiano cerasuolo
Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 28/02/2026
No matter the size or number, cats are always cats
Three pictures of cats.
Left: A domestic cat with leg over head, tongue out, washing itself, label is "GLM"
Middle: A tiger in same pose, label is "GLMM"
RIght: many cats together in same pose, label is "FIXED EFFECTS"
28313
Reposted by damiano cerasuolo
ArXiv Paperboy (Stat.ME+Econ.EM) @paperposterbot.bsky.social · 14/02/2026
arXiv📈🤖 BayesPPDSurv: An R Package for Bayesian Sample Size Determination Using the Power and Normalized Power Prior for Time-To-Event Data By
arXiv:2404.05118v1 Announce Type: new 
Abstract: The BayesPPDSurv (Bayesian Power Prior Design for Survival Data) R package supports Bayesian power and type I error calculations and model fitting using the power and normalized power priors incorporating historical data with for the analysis of time-to-event outcomes. The package implements the stratified proportional hazards regression model with piecewise constant hazard within each stratum. The package allows the historical data to inform the treatment effect parameter, parameter effects for other covariates in the regression model, as well as the baseline hazard parameters. The use of multiple historical datasets is supported. A novel algorithm is developed for computationally efficient use of the normalized power prior. In addition, the package supports the use of arbitrary sampling priors for computing Bayesian power and type I error rates, and has built-in features that semi-automatically generate sampling priors from the historical data. We demonstrate the use of BayesPPDSurv in a comprehensive case study for a melanoma clinical trial design.
011
Reposted by damiano cerasuolo
Heiko Breitsohl @hbreitsohl.bsky.social · 30/01/2026
A nice illustration of collider bias, in case anybody needs one. And it's about AI! What's not to love!? @dingdingpeng.the100.ci @p-hunermund.com
normaltech.ai
Fact checking Moravec's paradox
This famous aphorism is neither true nor useful
013
damiano cerasuolo @damiacer.bsky.social · 25/01/2026
(in the US. Where else?)
100
Reposted by damiano cerasuolo
Ken Koon Wong @kenkoonwong.bsky.social · 04/06/2025
I finally understood B-splines by working through the Cox-deBoor algorithm step-by-step, discovering they’re just weighted combo of basis functions that make non-linear regression linear. What surprised me is going through Bayesian statistics helped me understand the engine behind the model! #rstats
kenkoonwong.com
Understanding Basis Spline (B-spline) By Working Through Cox-deBoor Algorithm | Everyday Is A School Day
I finally understood B-splines by working through the Cox-deBoor algorithm step-by-step, discovering they're just weighted combinations of basis functions that make non-linear regression linear. What ...
2589
Reposted by damiano cerasuolo
Paul Hünermund @p-hunermund.com · 29/12/2025
On the plus side, LLMs also reduce barriers for non-native speakers, facilitate the discovery of prior literature, and remove traditional signals of scientific quality such as language complexity. www.science.org/doi/10.1126/...
science.org
Scientific production in the era of large language models
With the production process rapidly evolving, science policy must consider how institutions could evolve
3499
damiano cerasuolo @damiacer.bsky.social · 04/12/2025
Hélas, I can relate. I’ve never thought that my accent is “charmant”.
000
damiano cerasuolo @damiacer.bsky.social · 16/11/2025
A periodic reminder in #statistics:
statmodeling.stat.columbia.edu
How post-hoc power calculation is like a shit sandwich | Statistical Modeling, Causal Inference, and Social Science
040
Reposted by damiano cerasuolo
F. Javier Rubio @fjrubio.bsky.social · 23/10/2025
📘 An interesting initial book release by David Rossell on variable and model selection: 👉 davidrusi.github.io/modelSelecti... it provides accessible material for students learning the fundamentals of high-dimensional model selection, and it documents the R package modelSelection (formerly mombf).
davidrusi.github.io
High-dimensional model choice. A hands-on take
High-dimensional model selection with the modelSelection R package
073
Reposted by damiano cerasuolo
Jerry Coyne @evolutionistrue.bsky.social · 07/10/2025
Fred Ramsdell, who received the Nobel Prize for Physiology or Medicine yesterday for work on immunology, hasn't been informed of his win as he's "off the grid" hiking, and can't be contacted. www.theguardian.com/science/2025...
28914112
Reposted by damiano cerasuolo
Big Book of R @bigbookofr.com · 06/10/2025
Data visualisation using R, for researchers who don’t use R by Emily Nordmann, Phil McAleer, Wilhelmiina Toivo, Helena Paterson and Lisa DeBruine #RStats bigbookofr.com/chapters/data%20visu…
011
Reposted by damiano cerasuolo
Julia M. Rohrer @dingdingpeng.the100.ci · 02/10/2025
Just finished reading this *excellent* article by Gabriel et al. which discusses which effects can be identified in randomized controlled trials. With DAGs!> link.springer.com/article/10.1...
Elucidating some common biases in randomized controlled trials
using directed acyclic graphs

Although the ideal randomized clinical trial is the gold standard for causal inference, real randomized trials often suffer
from imperfections that may hamper causal effect estimation. Stating the estimand of interest can help reduce confusion
about what is being estimated, but it is often difficult to determine what is and is not identifiable given a trial’s specific
imperfections. We demonstrate how directed acyclic graphs can be used to elucidate the consequences of common imperfections,
such as noncompliance, unblinding, and drop-out, for the identification of the intention-to-treat effect, the total
treatment effect and the physiological treatment effect. We assert that the physiological treatment effect is not identifiable
outside a trial with perfect compliance and no dropout, where blinding is perfectly maintainedTable 1 showing the Identifiability of target estimands depending on whether there is blinding, full compliance, and no drop-outAn example DAG from the paper.
Fig. 4: A blinded trial with noncompliance.

U are unobserved confounders, Z is treatment assignment, C is compliance, X is the realized treatment, S is the subject's physical and mental health status, Xself and Xcln are the treatment that the participant and the clinician believed the participant received, Y is the outcome.
411424
Reposted by damiano cerasuolo
Simisani Nokulunga Ndaba @simisani.bsky.social · 22/09/2025
🚀 Curious about Positron, Posit’s new open-source IDE? This session shows RStudio users the basics, what’s new, and how it fits your workflow with @ivelasq3.bsky.social (Posit, PBC) #RStats #DataScience #Positron #rladies Thursday 09 October, 11am CT | 6:00 PM CAT www.meetup.com/rladies-gabo...
meetup.com
Positron for RStudio Users: A Gentle Introduction, Thu, Oct 9, 2025, 6:00 PM | Meetup
Curious about Positron, the new open-source data science IDE from Posit? This session will walk RStudio users through the basics of Positron, highlight familiar features, a
34319
damiano cerasuolo @damiacer.bsky.social · 19/09/2025
Europe thinks that free and private communication is not a fundamental right.
000
Reposted by damiano cerasuolo
Fight Chat Control @fightchatcontrol.bsky.social · 16/09/2025
Germany's position has been reverted to UNDECIDED. Despite expressing concerns about breaking end-to-end encryption, Germany refrained from taking a definitive stance on the Chat Control proposal during the September 12th LEWP meeting. A willingness to negotiate and compromise remains.
Overview of current support and opposition for the Chat Control legislation by Patrick Breyer (https://www.patrick-breyer.de/en/posts/chat-control/). Countries opposing include the Netherlands, Belgium, Luxembourg, Austria, Czech Republic, Poland, Slovakia, Finland, and Estonia. Undecided countries are Germany, Slovenia, Romania, and Greece. All remaining member states supports Chat Control.
9371344
Reposted by damiano cerasuolo
Chris Rackauckas @chrisrackauckas.bsky.social · 19/09/2025
DifferentialEquations.jl is many things, and lots of people only use a small portion of it. Check out the JuliaCon 2025 workshop: introduces many aspects of the packages that the developers feel are underutilized and under-understood! #julialang #sciml www.youtube.com/watch?v=lSGF...
youtube.com
A Deep Dive Into DifferentialEquations.jl | JuliaCon Global 2025 | Rackauckas, Smith
YouTube video by The Julia Programming Language
0113
damiano cerasuolo @damiacer.bsky.social · 19/09/2025
catholic church: always on the right side of the history 🙄 [pope leo XIV says that the holy see doesn’t pronounce the word g3nocid3]
010
Reposted by damiano cerasuolo
Max-Planck-Gesellschaft @maxplanck.de · 21/08/2025
The anti-autocracy handbook: how scientists can cope with democratic backsliding. This guide is a call to action, resilience, and collective defence of #democracy and #academicfreedom in the face of mounting authoritarianism zenodo.org/records/1569... @EU_Commission
03018
damiano cerasuolo @damiacer.bsky.social · 02/06/2025
Bayesian Hierarchical Mixture Cure Modelling Framework to Utilize Multiple Survival Datasets for Long-Term Survivorship Estimates #stats
onlinelibrary.wiley.com
A Bayesian Hierarchical Mixture Cure Modelling Framework to Utilize Multiple Survival Datasets for Long‐Term Survivorship Estimates: A Case Study From Previously Untreated Metastatic Melanoma
Time to an event of interest over a lifetime is a central measure of the clinical benefit of an intervention used in a health technology assessment (HTA). Within the same trial, multiple end-points m...
010
Reposted by damiano cerasuolo
Julian Hoffman @julianhoffman.bsky.social · 09/04/2025
Any potential gain from a switch to renewables and tech-heavy futures is inevitably undone if the underlying system of extractivist colonial enterprises and growth at all costs remains unchanged. The suffering just gets shifted to new places. www.theguardian.com/environment/...
theguardian.com
Revealed: Big tech’s new datacentres will take water from the world’s driest areas
Amazon, Google and Microsoft are building datacentres in water-scarce parts of five continents
16725
damiano cerasuolo @damiacer.bsky.social · 28/03/2025
AI’s anwers are still so biases and prompt-dependent, I don’t see how it could replace humans. www.cnbc.com/2025/03/26/b...
cnbc.com
Bill Gates: Within 10 years, AI will replace many doctors and teachers—humans won't be needed 'for most things'
A new era of "free intelligence" powered by AI will change the way humans work, says billionaire Microsoft co-founder Bill Gates.
020
damiano cerasuolo @damiacer.bsky.social · 21/03/2025
#r code written by AI seems to be unnecessarily complicate.
000
Reposted by damiano cerasuolo
Steven Ponce @sponce1.bsky.social · 18/03/2025
📊 #TidyTuesday – 2025 W11 | Palm Trees . 🔗: stevenponce.netlify.app/data_visuali... . #rstats | #r4ds | #dataviz | #ggplot2
Alluvial diagram showing growth form patterns across palm subfamilies. The visualization flows from left to right, connecting palm subfamilies (Arecoideae, Calamoideae, Ceroxyloideae, and Coryphoideae) to growth forms (Erect, Non-erect, and both) to stem types (Solitary Stem, Clustered Stems, and both). The diagram reveals distinctive patterns: Arecoideae palms show diverse growth habits, Calamoideae are predominantly erect with clustered stems, Ceroxyloideae show a mix of growth forms, and Coryphoideae are mainly non-erect with solitary stems. The flow width represents the percentage of species with each characteristic.
0152
damiano cerasuolo @damiacer.bsky.social · 11/03/2025
bioRxiv and medRxiv will be managed by a new organization called openRxiv. www.nature.com/articles/d41...
nature.com
Preprint sites bioRxiv and medRxiv launch new era of independence
The popular repositories, where life scientists post research before peer review, will be managed by a new organization called openRxiv.
010
Reposted by damiano cerasuolo
Wolfgang Viechtbauer @wviechtb.bsky.social · 06/03/2025
The Evidence Synthesis and Meta-Analysis in R Conference (ESMARConf) is back! It will be held June 11th to the 13th, 2025: esmarconf.org/2025/ Recordings of the talks and workshops from previous years can be found here: esmarconf.org/recordings/ #ESMARConf #MetaAnalysis #EvidenceSynthesis #RStats
esmarconf.org
2025 - ESMARConf
0139
Reposted by damiano cerasuolo
Crystal Lewis @cghlewis.bsky.social · 04/03/2025
Just a reminder for anyone who receives SPSS, Stata, or SAS files but does not have licenses for these programs (🙋), you can import any of these files into #rstats and work them! My favorite package for working with these files is {haven}. haven.tidyverse.org
311322
Reposted by damiano cerasuolo
Joey Politano🏳️‍🌈 @josephpolitano.bsky.social · 02/03/2025
Now that we've cut all US foreign aid, we can at least spend that money on helping people in need here at home, for example by gambling on crypto to support our Great Patriot Bag Holders
421820
damiano cerasuolo @damiacer.bsky.social · 27/02/2025
study led by researchers from Oxford Population Health has shown that environmental factors (i.e. smoking and physical activity) and living conditions have a greater impact on health and premature death than our genes. #health #medicine www.nature.com/articles/s41...
nature.com
Integrating the environmental and genetic architectures of aging and mortality - Nature Medicine
Based on a systematic analysis of environmental exposures associated with aging and mortality in the UK Biobank, the relative contributions of such exposures and genetic risk for mortality and a range...
000
Reposted by damiano cerasuolo
New Scientist @newscientist.com · 27/02/2025
Feedback is intrigued by a theory arguing that when maleness is threatened, men overcompensate with increased support for war and homophobia – and interest in SUVs
newscientist.com
How a theory about maleness could explain the state of the world
Feedback is intrigued by a theory arguing that when maleness is threatened, men overcompensate with increased support for war and homophobia – and interest in SUVs
267
Reposted by damiano cerasuolo
Olivier Simard-Casanova @o.simardcasanova.net · 16/02/2025
Due to the poor productivity of the time, our ancestors also had to use children as free labor Should we bring back child labor, then? These people may claim to have the highest IQ or be “geniuses” and whatnot, but they're actually just ignorant, uncurious, and overconfident
55014
Reposted by damiano cerasuolo
Philip Gourevitch @pgourevitch.bsky.social · 15/02/2025
What these nihilistic vandals are doing to dismantle science & medicine—generations of expertise & public service & life improving benefits—is cultural revolution level national self harm. No foreign foe could inflict such brutal damage & senseless suffering on US so fast & effectively & unresisted.
476179815648
damiano cerasuolo @damiacer.bsky.social · 13/02/2025
If you ask #LeChat by #MistralAI whether it uses or not your data, it answers that it is not concerned since it is not associated with Mistral AI and its application. But questions about its data use have been raised. To read [in French] www.frandroid.com/culture-tech...
le chat answer to data use
010
Reposted by damiano cerasuolo
Solomon Kurz @solomonkurz.bsky.social · 04/02/2025
New blog up: solomonkurz.netlify.app/blog/2025-02... This time I dip my toes into causal inference for quasi-experiments using matching methods, and my use case has missing data complications. Many thanks to @dingdingpeng.the100.ci and @noahgreifer.bsky.social for their peer review! #RStats
solomonkurz.netlify.app
Matching, missing data, a quasi-experiment, and causal inference--Oh my! | A. Solomon Kurz
I'm finally dipping my does into causal inference for quasi-experiments, and my first use case has missing data. In this post we practice propensity score matching with multiply-imputed data sets, and...
1311934
damiano cerasuolo @damiacer.bsky.social · 04/02/2025
#Rlinguo, the #R #rstats app on mobile phones. (And it does not collect your data ;)) rtask.thinkr.fr/rlinguo-why-...
rtask.thinkr.fr
Rlinguo — Why Did We Build It? - Rtask
The R task Force - R experts for all your needs
051
damiano cerasuolo @damiacer.bsky.social · 29/01/2025
Is it adjusting for baseline covariates in clinical trial analyses like fishing for statistically significant results? #r #rstats #clinicalresearch #statistics www.r-bloggers.com/2025/01/thin...
r-bloggers.com
Thinking about covariates in an analysis of an RCT
I was recently discussing the analytic plan for a randomized controlled trial (RCT) with a clinical collaborator when she asked whether it’s appropriate to adjust for pre-specified baseline covariates...
053
Reposted by damiano cerasuolo
Mark Rieke @markjrieke.bsky.social · 13/01/2025
evergreen (currently at ~6hrs of training 🫡)
032
damiano cerasuolo @damiacer.bsky.social · 10/12/2024
#onthisday in medicine. #HIV #MedSky
010
Reposted by damiano cerasuolo
Dax Kellie @daxkellie.bsky.social · 09/12/2024
If you missed my talk but still want to learn how to make your R code more reproducible, my slides are here 🙂: daxkellie.quarto.pub/will-your-co... All the links to packages and resources I mentioned are there, so hopefully this can be a nice reference, too! #ESAus2024 #rstats #quartopub 🧪🌏
Title slide: Will your code run again? Tips for making code reproducible in R
15347107
damiano cerasuolo @damiacer.bsky.social · 08/12/2024
Decision curve analysis (DCA) combines classical assessment of predictive performance with the consequences of using these strategies for clinical decision-making. #Bayesian DCA allows for an intuitive probabilistic interpretation framework, while incorporating prior evidence. #rstats #statistics
onlinelibrary.wiley.com
Bayesian Decision Curve Analysis With Bayesdca
Clinical decisions are often guided by clinical prediction models or diagnostic tests. Decision curve analysis (DCA) combines classical assessment of predictive performance with the consequences of u....
030