How do you keep the music playing? youtu.be/eeK1xc418S4?si… via YouTube - one of the most beautiful things I’ve heard in a long while. Helen Charlston Trinity College Choir Cambridge

Randomisation in 3 steps.
1) what is seen trumps probabilities of what might have been seen
2) for the unseen, probabilities must do
3) randomisation helps with 2).

Myths of randomisation growkudos.com/publications/1… via @growkudos

I've updated my template for Quarto statistical reports using R: hbiostat.org/rr/#template #rstats #Statistics . Thanks to Vanderbilt Department of Biostatistics Max Rohde for starting this template and introducing me to Quarto

Judea Pearl Frank Harrell Peter Tennant Boris Sobolev One of the joys of this platform is that people who confuse randomisation & random sampling, don’t understand study effects or concurrent control or informed consent, confuse ICH & FDA & have probably never designed, run or analysed an RCT can lecture you on them.

Ken Burns warns about Trump during his commencement speech at Brandeis University: so powerful, so wise, so emotional, so true. It shook me to my core. Tearing up a river.

Nice to see continuous variables as trial endpoints for a change! But analyses will be disturbed by transformation uncertainty & covariate adjustment is needed. Calls for semiparametric ordinal models which can also handle clinical event interruptions: hbiostat.org/ordinal

Yes! Richard Riley (R²) is doing great things for the Biostat and clinical research communities

Beautiful example of corporate non-greed

datamethods is a great place for brainstorming and giving feedback about statistical analysis plans. I've put some of my thoughts there and hope that many others will add theirs.

“As with political parties, what’s striking to me is not so much that some scientists bullshit, but that the mainstream of the scientific establishment doesn’t seem to care.”
statmodeling.stat.columbia.edu/2024/05/05/on-…

Today the annual 4-day virtual Regression Modeling Strategies course begins. We always have a wonderful array of actively engaged participants and this year is no exception. #rmscourse hbiostat.org/doc/rms/4day

Fantastic to see two new studies from Rasha Al-Lamee's ORBITA programme, presented at #EuroPCR and simultaneously published in JACC Journals. ORBITA2 found PCI was an effective anti-anginal procedure. However, do all patients benefit similarly? How important is the nature of symptoms?

Two placebo-controlled studies with a synergistic outcome both published in JACC by two fabulous fellows from the ORBITA team Christopher Rajkumar Florentina Simader

Collaborating with researchers who are bold enough to be truly innovative in clinical research including being all-in on Bayes is an absolute joy. Go ÓRBITA team!! Rasha Al-Lamee Matthew Shun-Shin

Registration is about to close on this year's 4-day Regression Modeling Strategies Course which starts on Thursday: hbiostat.org/doc/rms/4day #rmscourse #Statistics

Updated price list:

Registration is about to close for tomorrow's Pre-Regression Modeling Stategies 1-day course introducing R, linear models, and Quarto: hbiostat.org/doc/rms/1day.h… #rmscourse Vanderbilt Department of Biostatistics