@f2harrelli
iAccount based inUnited States
About this account
- Account based in
- United States
- Connected via
- Web
Account-level information from X, not a live location or the device used for a specific post.
Biostatistician/Professor/Consultant/Founding Chair of Biostat, Vanderbilt U. Blog: Statistical Thinking:https://nitter.cf/t.co/2BTEONzsfX @f2harrell on https://nitter.cf/t.co/bsPN9JQNOS
Nashville, TN
Joined January 2017
- Tweets19.8K
- Following174
- Followers30.4K
- Likes34.9K
Frank Harrell retweeted
*BRILLIANT*
The French made a short film about what could happen after all immigrants would follow the ‘_white, angry old man’s_’ advice to go back where they came from.*
Believe it or not these are all special cases of the cumulative probability family of ordinal regression models as discussed in fharrell.com/post/cpm #Statistics
One R function, orm (in the rms package), will fit every model except when using random effects for paired data. The blrm function in the rmsb package handles that.
I've been going back and forth between @zeddotdev and @CotEditor and have found that I prefer Cot when not using AI. For the one language I needed syntax highlighting for (@typstapp ) that was not built-in to Cot, Claude quickly wrote the Cot plug-in needed.
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
Frank Harrell retweeted
It’s astonishing that someone has spent so much effort attempting to write about RCT through a causal inference lens while spending so little effort learning *literally anything else* about the design, execution, or analysis of clinical trials
Frank Harrell retweeted
Replying to @yudapearl
"Data science is statistics without a plan." — old proverb
Frank Harrell retweeted
In 1988 Dolly Parton pulled every 7th and 8th grader in Sevier County into a room and told them to pick a buddy.
She made them sign a contract.. if BOTH graduate, she hands you both $500. One fails, nobody gets paid.
The county dropout rate went from 35% to 6%.
She handed out the checks herself.
In 1989 she offered $500 to every kid in the county who wanted to go to the local college.
When Gatlinburg burned in 2016, she sent 900 families $1,000 a month for six months.
You've probably heard that part.
Here's the part you haven't:
The fund raised more than expected, so she showed up in May and handed every one of those families another $5,000 on top.
Then she quietly put $3 million into the recovery group that handles the years after the news trucks leave.
She built her county an entire hospital.
She played a benefit concert in 2007, Dollywood and the Stampede matched it, and LeConte Medical Center opened in 2010 with a birthing unit and a women's center in it.
In 2022 she started covering 100% of tuition, books, and fees for Dollywood employees.
Full-time, part-time, AND seasonal.
In 1991 she paid for a bald eagle sanctuary because the birds couldn't be released back into the wild, no press release.
And when Hurricane Helene came through here in Western North Carolina, she wrote a personal check for $1 million and her companies matched it.
She didn't owe North Carolina anything, but she helped us anyway.
She started the Imagination Library because her daddy couldn't read or write. Before he passed, he told her it was the most important thing she ever did.
Rest in Peace, Dolly 🇺🇸
The gold standard for producing PDF documents (especially books and journal articles) has been LaTeX. @typstapp re-engineered the entire process in an incredibly elegant extensible way. On my MacOS LaTeX (TexLive) takes 9.1GB of disk space. Typst takes 43MB. Not a misprint.
Frank Harrell retweeted
Replying to @DrewMoghanaki
Reminds me of this: ‘Most users of propensity score matching do not even entertain the notion that the treatment effect may interact with propensity to treat, must less entertain the thought of individual patient characteristics interacting with treatment.’ @f2harrell Ch 17
Frank Harrell retweeted
Ron Carter and Herbie Hancock. August 19th, is the day, ladies and gents... The day that I shall be on stage to find THOSE notes with none other than Mr. @herbiehancock at the @HollywoodBowl celebrating 100 years of @milesdavis .
Find those tickets: RonCarterJazz.com
I appreciate @BaimInstitute making this video available. This was fun, and I hope the talk's content will stimulate others to think about how such approaches can make trials more efficient and will stimulate further discussion. There's plenty of controversy to go around.
The recording from Baim Grand Rounds with @f2harrell on "Modernizing Clinical Trial Design and Analysis to Improve Efficiency & Flexibility" is now available on the Baim website. You can find it here: baiminstitute.org/index.php/…
Is it worthwhile to parametrically fit a distribution? Or better to just go with the empirical cumulative distribution function (ECDF) and its generalization - cumulative probability ordinal regression? New article: fharrell.com/post/ecdf #Statistics
So many misunderstandings about clinical trials, causal inference, denominators, and personalized/precision medicine all rolled into one.
❗️ An article with multiple conceptual errors; “randomization is not magic”, “chance alone can produce important imbalances” = false , see jclinepi.com/article/S0895-4…
Frank Harrell retweeted
Reminder! Join us on Monday, August 10, 2026, from 9:30-10:30 am EST for Baim Grand Rounds with @f2harrell, presenting "Modernizing Clinical Trial Design and Analysis to Improve Efficiency & Flexibility.” Zoom link: us06web.zoom.us/j/8570298937…
Frank Harrell retweeted
The Wee P Boogie: The Song That's Sweeping the Nation!
Send this snappy tune to all P-value users.
youtube.com/watch?v=PnW39ywW…