New presentation: www.fharrell.com/talk/qrct/ #Statistics #StatsSky #clinicaltrials

In ML, everyone’s Humpty Dumpty
statmodeling.stat.columbia.edu/2026/04/17/i...

Right, the old fashioned way. My goal was to have a tool that gave me great hard copy to debug with when I don't have my laptop.

bat look extremely helpful. Thanks for the pointer. Looks like to print I'd use bat with prettybat, but I don't see R listed as a supported language.

There is also an "st" command line option to get a symbol table at the end of the R code listing. #RStats

Color-highlighted pretty-printing of R code

Python script for pretty-printing of R code files: github.com/harrelfe/rsc...

Was Cot Editor and I use RStudio when I need to debug R code especially when inside Quarto scripts.

Its extensions for R and Quarto don't have the interactivity you're seeking but Zed is extremely extendable and I'll bet someone will work on full Quarto, R, and other IDEs with Zed.

It's based on the same code model as Visual Studio which many developers have found to not scale efficiently to large codebases. But I'll bet it has some great features.

Cot is my 2nd favorite. It's fantastic. Zed is a bit faster, has nicer simpler layout with better default fonts, and is built for AI.

Zed — Love your editor again Zed is a high-performance, multiplayer code editor from the creators of Atom and Tree-sitter.

I have a new favorite editor/IDE: Zed. Open-source, blazing fast, built-in AI operability, easy to use. Heard that lots of developers are dropping Visual Studio Code for Zed. Has R and Quarto extensions. zed.dev

A fantastic reproducibility autopsy that makes me even more dubious of both microbiome & autism research than I was previously & more sickened by how scientific journals work. For them to make money capitalizing on free reviewer labor the least they can do is to be scientific quality gatekeepers.

Screenshot of the announcement by Simon Urbanek regarding Tomáš Kalibera’s untimely passing. It is with great sadness that we announce the passing of Tomáš Kalibera on 1 April, a valuable member of the R core team for close to 10 years and a good friend, after a short but aggressive illness. Tomáš brought in fresh perspective and knowledge, enabling him to improve many aspects of R, including performance and reliability. He created many tools aimed at aiding package authors to make their packages more reliable, and was instrumental in modernizing the Windows build of R. He was an active member of CRAN and the R community, providing help to package authors, and he was the most prolific writer on the R core blog. He will be remembered for his profound contributions to R by millions of users. He is survived by his wife and 1 year old son. A full obituary will be posted in due time. Respectfully, Simon Urbanek

RIP Tomáš Kalibera. #rstats lost a huge contributor today. Condolences to his young family.

stat.ethz.ch/pipermail/r-...

Donald Trump was posting on Truth Social until 4:10am, including an image of himself as Jesus and multiple posts of the same article, raising alarms about Biden’s cognitive decline.

I've recently made some changes to the gallery of all my data visualisations 📊

Most examples now have links to the underlying source code! If you see something you like, you can see how it was made 🤩

#DataViz #RStats

But as an Alabama native I can say that Alabama makes up for poor education, poor social services, poor health care, poor race relations, poor environmental policy, and repressive politics by having great football.

A huge point about meta-analysis, which begs for reliance on individual patient-level data and adjusting for a full set of covariates.

Also consider our creating a new special topic on datamethods.org for which I could assign a short URL e.g. datamethods.org/ltte . This would allow for many others to add their own unpublished LTTEs.

Causal by Design: new blog article on Statistical Thinking. fharrell.com/post/causal/ #Statistics #StatsSky #EpiSky

Interesting read. Would be curious to hear other takes. Also, what about REMAP!? 😀

cc @andygarrett.bsky.social @marionkcampbell.bsky.social @stephensenn.bsky.social @vcornelius.bsky.social @f2harrell.bsky.social @gsmaclennan.bsky.social

This is an an excellent article and completely matches my experience. Many guidelines are stifling innovation and not leading to better trials and CROs are absorbing $$$. The article fails to mention the huge gains that can come from fully sequential Bayesian trials with no fixed sample size.

Yes all sessions are recorded and are available to participants by the next day. In the past many UK/Europe/Africa participants have participated live for the first 1/2 of each day and watched videos for the 2nd half when they wake up.

1-day online course May 11: Intro to R/RStudio & an efficient comprehensive R workflow, introducing the R rms package, enhancing regression analysis skills to be ready for the RMS 4-day course. Details and registration at hbiostat.org/course @amstatnews.bsky.social @instats.bsky.social #RStats

Regression Modeling Strategies 4-day virtual course May 14-15, 18-19. More than regression:principled flexible modeling, accuracy, interpretation, selection, stability, validation, Bayes, contrast w/ML &more. hbiostat.org/course @amstatnews.bsky.social @instats.bsky.social #Statistics #StatsSky

19  Diagnosis – Biostatistics for Biomedical Research

hbiostat.org/bbr/dx. Related to the great statistician Dawid's article explaining why probability of disease is the diagnostic problem while sensitivity & specificity are part of the "sampling model". Sens & spec are only consistent with outcome-based sampling.

I remember the era where ultra-wealthy had to share the burden of supporting democracy. No longer, unfortunately. The decline of the U.S. started with Reagan, and Newt Gingrich's (U.S. House republican) contract on America. Trump has taken this to unimagined new levels.

Weekend reads: Half of social science ‘doesn’t replicate’; ‘Scientific ghosts: Life after retraction’; multisensory learning paper retracted If your week flew by — we know ours did — catch up here with what you might have missed. The week at Retraction Watch featured: A citation alert led researchers to a network of fake articles. But w…

Weekend reads: Half of social science ‘doesn’t replicate’; ‘Scientific ghosts: Life after retraction’; multisensory learning paper retracted

As an aside, a case-control study is the only design where sensitivity and specificity make any sense.

I think there is a difference between the probabilities of the two events. That's why I don't like to use 'false positive' in medical probabilistic diagnosis but rather P(disease | test results, background variables).

But does 'false positive' mean a positive that turns out to be false, or a true negative that turns out to be positive? The language isn't clear.