Important message below but infuriating #AI field has to 'discover' this rather than being guided by long-standing stats literature

eg, onlinelibrary.wiley.com/doi/10.1002/si…
by Gary Collins 🇪🇺 says:
'Categorising continuous predictors is ... inefficient & should not be used in model development'

Teaching or learning stats and R?

I'm creating self-graded practice exercises so that students can test their understanding of the material without the help of an instructor.

I just created some related to how to estimate causal effects using randomized experiments ...

#rstats

Interested in #models for multi-state #survival #data ? Then this book might be for you! Find more details here: routledge.com/Models-for-Mul… and make use of 20% off with the discount code IBS20 (valid until 31st March, can't be used with other offers). #rstats CRC Press Data Science, Stats, and Math Books

Don't let the causal inference buzz fool you: Description is the foundation of science.

We've described the 3-year health impact of #COVID19 in Madrid, the EU region with the highest life expectancy.

If we can't describe, no #causalinference can follow.
academic.oup.com/ofid/article/1…

Like Bayesian stats & causal inference? Like secret BONUS content? Some of my Stat Rethinking lectures have BONUS topics hiding at end: misclassification, survival analysis, poststratification, Mundlak machines, Cholesky factors, floating point monsters. Thread w timed links:

#MethodMonday 48: Estimating means/SDs from medians/IQRs/ranges?
1) doi.org/10.1186/1471-2…
2) training.cochrane.org/handbook/curre…, sections 6.5.2.9 + 10.5.3, 6.5.2.4-6

This week covers methods for estimating means and SDs from medians/IQRs/ranges, as frequently done in meta-analyses. 1/..

Do you use logistic regression? If so, you’ll want to read the thread below.

⚠️ Warning: Memes, charts, #rstats , and practical advice ahead.

Delighted to collaborate with Australian paed nursing research colleagues
Mari Takashima Amanda Ullman
Perth Children's Hospital @queenslandchildrenshospital @royalchildrenshospital Curtin University

What is the harm using paediatric invasive devices? Find out here:
nature.com/articles/s4139….

An amazing editorial on semi-futility of precision #Oncology . Would love to get Stephen John Senn ‘s take. I would rather see a treatment have a big average effect then find out who it doesn’t work on. #personalized

Causal Inference Is Not Just a Statistics Problem. Lucy D’Agostino McGowan, Travis Gerke & Malcolm Barrett. Journal of Statistics and Data Science Education. tandfonline.com/doi/full/10.10…

Statistical #PowerAnalysis currently dominates #ExperimentalDesign . In this Essay, Shinichi Nakagawa | 中川震一 &co argue that we should move away from the current focus on power analysis and instead encourage smaller scale studies & collaborative projects #PLOSBiology plos.io/48Gk8Og

Confidence distributions are distributional summaries of evidence. They’re like Bayesian posteriors but without prior distribution assumptions. I wrote a tutorial paper in Statistics in Medicine for clinical trial statisticians AusTriM NHMRC CTC
onlinelibrary.wiley.com/doi/10.1002/si…

Was rereading one of my favorite papers of all time today and if this isn't one of the best table 1s ever created, I don't know what is

It is 2024 and it is still common practice in the medical literature to put a bunch of covariates in a regression model for some bad health outcome and declare the significant associations as 'risk factors' that people should worry about. Pseudo science