Robin Blythe
banner
rbly.bsky.social
Robin Blythe
@rbly.bsky.social
Assistant professor at Duke-NUS medical school. Mostly interested in health economics, biostats and clinical informatics.
Basically, fit the model with chains = 0, then create an update() function with recompile = FALSE and the sampling specs you want. I've usually had success with {{furrr}} for the parallelisation part. Can't take full credit for this one as I needed an LLM to show me how the recompile step worked.
November 20, 2025 at 1:10 PM
Not even yourself
November 14, 2025 at 3:52 AM
Absolutely loving A City on Mars by @zachweinersmith.bsky.social and @weinersmith.bsky.social but feel the need to point out an error in the text: Kelly and Zach appeared to have accidentally reversed the axis on this figure
October 28, 2025 at 9:40 AM
When studies like the below are publicised, the reaction from technical experts is heavy scepticism. By contrast most c-suite, mgmt consultants, and anyone with "Founder" in their bio breathlessly proclaims a new era of medicine, and they're WAY louder.
Found on LinkedIn, naturally
#statsky #medsky
October 1, 2025 at 3:20 PM
Reminds me of this classic:
September 26, 2025 at 3:53 PM
Me: *waiting for my hideously large dataset to be analysed*
My brain: "I wonder if 2 Chainz is a Bayesian"
August 28, 2025 at 3:25 PM
Our Lancet Haematology commission has been published😎
doi.org/10.1016/S235...
My team's section (4) modelled what global development targets SHOULD be, rather than the current targets we've set and will fail to achieve by 2030. I built a model to inform future anaemia goals. #healtheconomics #rstats
August 27, 2025 at 2:30 AM
I've found that the best way to respond to scam messages is to speak solely in made up Chinese proverbs
August 21, 2025 at 4:14 AM
Thou shalt, if thou writest a medicinal treatise applying thine chosen algorism of #machinelearning or #artificialintelligence, use the word "leveraging" in the title to make manifest the newness and preeminence of thy research.
AMEN.
August 4, 2025 at 1:26 AM
Kinda always wanted this shirt from @theonion.com for similar reasons
August 3, 2025 at 1:58 PM
Tired: Huh, bottomless portafilters are really cool for learning about espresso

Wired: The distribution of coffee coming out kinda looks like a 3D trace plot. I wonder if I can write a MCMC program that tells me how to get the most even extraction
July 28, 2025 at 6:51 AM
Light holiday reading courtesy of
@rmcelreath.bsky.social
July 19, 2025 at 7:22 AM
June 13, 2025 at 10:14 AM
How I felt after making this thread
June 2, 2025 at 9:28 AM
Babbage inventing his difference engine (1830, colourised)
April 2, 2025 at 2:34 PM