New paper with Maarten van Smeden and Tra My Pham: The marginality principle revisited: Should “higher-order” terms always be accompanied by “lower-order” terms in regression analyses? I love Biom J for including this figure without batting an eyelid! 1/ onlinelibrary.wiley.com/doi/10.1002/bi…

📢 Our latest research on vaccine effectiveness is now available at arxiv.org/abs/2310.04578. 🧪💡Our contributions include the efficiency theory under biased TND sampling, a one-step, doubly robust, and locally efficient estimator implementable with machine learning.

The prize for most original poster at #AEET2023 so far definitely goes to iBartomeus et al for this amazing effort!

What should you do when your simulation study results aren't what you expected?🤔 Following advice from Ian White, Tra My Pham, Matteo Quartagno & Tim Morris can help you avoid errors and improve the quality of your studies. Read now in IJE👇bit.ly/3QiIwiS

Mark your calendars! The 16th #KolokotronesSymposium takes place 12/15 in-person at Harvard T.H. Chan School of Public Health and virtually. Join us for a CAUSALab & Harvard Perinatal & Pediatric Pharmacoepidemiology event. Learn about #causalinference in #pregnancy. Registration opens soon - stay tuned!

We reviewed 100 psych. simulation studies & find room for improvement in planning/reporting. As a remedy, we (František Bartoš, Tim Morris, Anne-Laure Boulesteix, Daniel Heck (@[email protected]) & Samuel Pawel) present ADEMP-PreReg, a sim study preregistration & reporting template osf.io/ufgy6/

Calculus Made Easy is a classic and elegant book on calculus originally published in 1910 by Silvanus P. Thompson! And now it has a free online version! abakcus.com/tool/calculus-…

"Draw your assumptions before your conclusions" 5 years ago we launched the *free* #CausalDiagrams course Harvard Online @edxOnline. Since then, 80,000 people in 180 countries have registered. Check it out if you want to learn about DAGs and SWIGs : edx.org/learn/data-ana…

Three vaccines (MMR, Flu, and HPV) and their risks, compared against becoming infected with the disease itself. nytimes.com/2020/01/09/opi…

Good news TRIPOD+AI #reporting recommendations for prediction model studies using either regression or #machinelearning methods has been accepted for publication 🥳🎉 #ArtificialInteligence #transparency #reportingstandards #standardsforAI

All 3 papers (#openaccess) now out in the The BMJ on the **Evaluation of clinical prediction models** part 1: tinyurl.com/3h4zfvfw part 2: tinyurl.com/ynytrpmx part 3: tinyurl.com/2cfc9uj3 #stats #MachineLearning #predictiveanalytics

Testing for assumptions is not great practice. The test will lead to model changes if the assumption is violated, ruining the validity of uncertainty quantitfication procedures (e.g., CIs) We should use models with assumptions that are reasonably known to be correct a priori.

Do machine learning methods lead to similar individualized treatment rules? A comparison study on real data. Florie Bouvier, Etienne Peyrot, Alan Balendran, Corentin Ségalas, Ian Roberts, François Petit, Raphaël Porcher. Statistics in Medicine. onlinelibrary.wiley.com/doi/10.1002/si…

it's extremely powerful when all your projects fall under one overarching goal such that they feed into and enhance each other.

CFIRB: Soumission d'abrégés pour les affiches jusqu'au 30 avril -- plusieurs prix de voyage à gagner ! / Submit a poster abstract by April 30th -- Several travel awards to win! crmath.ca/activites/#/ty…

In 1599 the church used a placebo controlled trial to test if a French girl was possessed by a demon. Holy objects and identical, non-blessed objects were shown to the girl. "She reacted similarly when exposed to both genuine and sham religious objects" journals.sagepub.com/doi/full/10.11…

PSA: All #rstats package on #cran will get an official DOI! This will facilitate bibliometrics and giving credit to R package authors. Registering all 20,000+ packages will still take a few more days. But the first couple of thousand are already live. Example:

The groundbreaking paper demonstrating that neural networks are often miscalibrated and tend to produce overconfident predictions has already garnered over 6,000 citations. Despite this, some still occasionally make ludicrous claims on social media that 'calibration doesn’t