New paper by @Lester_Domes and me. We discuss why uniformity is central to the validity of P-values and why some Bayesian variants don’t meet this, other units for S-values besides base-2 logs, and relation of S-values to other stat measures of information

Whoa—my book is up for pre-order! Model to meaning: How to interpret Statistical & ML Models in #RStats and #PyData The book presents an ultra-simple and powerful workflow to make sense of ± any model you fit Note: The web version stays free forever. https://t.co/fpDmSdDWov

My notes from teaching this year’s module on ‘inverse problems & learning from data’. Mainly inverse problems view but tried to also link to ML, stat etc. Some matrix calc, splines with physics-informed regularisation, tiny bit of neural nets etc in there… https://t.co/JNoL0FRK1H

Imagine consistent/intense rage bait being one of the biggest threats to people’s health assuming this study is methodologically/statistically sound (a giant assumption to make) and that what it’s attempting to infer is actually true

Truly the gift that keeps on giving

The irony being he then spent the end of his career trying to make sure that was the case. You either die a hero or live long enough to see yourself become the villain.

Happy to share our new article "Common wrong beliefs about statistical testing: recent trends in biomedical sciences". @Mohamma70696197 https://t.co/bFLHCF17kl

I’m excited to announce my new book FOOD INTELLIGENCE coauthored with brilliant journalist @juliaoftoronto! Cutting through the confusion about diet & health, #FoodIntelligence describes the fascinating science of how food both nourishes and harms us. https://t.co/xQHklDYVhy

@learnfromerror @david_colquhoun @vamrhein @dailyzad Most readers rely on common word meanings to understand results, and so mistake "statistical significance" for practical importance. Compatibility is a logically weaker term which suggests nothing about importance. For details see Rafi-Greenland and

Paper: https://t.co/eRA1Y03vtZ Blog post:

One year ago today... 60 hours that forever changed hip-hop history 💣 Favorite track here?

After 21 years at my dream job, I’m very sad to announce my early retirement from the National Institutes of Health. My life’s work has been to scientifically study how our food environment affects what we eat, and how what we eat affects our physiology. Lately, I’ve focused on

@umichkim Just gonna note that imo the work posted is based on this prior work and the behind the scenes work that went into it but has not credited this appropriately for reasons having nothing to do with proper citation practices

Here’s a thing. A bit bitter sweet for a few reasons, and also my name is spelled wrong lol but still

#stats reporting idea: I was reading Greenland et al. (2016) and they mention reporting p-values for effects of interest (as compatibility measures) alongside the null. Seems like an interesting idea. Plot is for 0, .1, .3, .5. https://t.co/BESAgXUMZX

Rest in peace, Ed Leamer

RT @analisereal: Ed was a great, researcher, mentor and human being. It's hard to describe how impactful he was to my formation---he would…

at the current rate, the asteroid will have a 1,200% chance to hit us. it will hit us, on average, twelve times.

it would he good for everyone's hearts to see izzy and alex pereira training together

Take it from someone who had this role for a couple of years: this is legitimately an interesting and rewarding role for an early-mid career statistician in medicine. Not sure if anyone is still reading tweets, but RTing for whatever reach I can offer…

Tutorials in Biostatistics Multiple Imputation for Longitudinal Data: A Tutorial. Rushani Wijesuriya, Margarita Moreno-Betancur, John B. Carlin, Ian R. White, Matteo Quartagno, Katherine J. Lee. Statistics in Medicine.