Anne-Laure Boulesteix
@boulesteixlaure.bsky.social
Statistician and metascientist. Professor of biometrics at LMU Munich Medical and Mathematical Faculties, committed to open science, member of the Munich Center of Machine Learning. Opinions are mine.
Reposted by Anne-Laure Boulesteix
Comment by the STRATOS initiative: (including @boulesteixlaure.bsky.social @mhstat.bsky.social) onlinelibrary.wiley.com/doi/10.1002/... (6/8)
Commentary: Regression Models—Efforts Are Required to Improve Statistical Practice and Teaching
Click on the article title to read more.
onlinelibrary.wiley.com
June 26, 2025 at 12:28 PM
Comment by the STRATOS initiative: (including @boulesteixlaure.bsky.social @mhstat.bsky.social) onlinelibrary.wiley.com/doi/10.1002/... (6/8)
For researchers with limited background in statistics, we have written this gentle introduction to simulation studies (with @timpmorris.bsky.social and other people from the STRATOS initiative):
bmjopen.bmj.com/content/10/1...
bmjopen.bmj.com/content/10/1...
Introduction to statistical simulations in health research
In health research, statistical methods are frequently used to address a wide variety of research questions. For almost every analytical challenge, different methods are available. But how do we choos...
bmjopen.bmj.com
June 23, 2025 at 8:37 AM
For researchers with limited background in statistics, we have written this gentle introduction to simulation studies (with @timpmorris.bsky.social and other people from the STRATOS initiative):
bmjopen.bmj.com/content/10/1...
bmjopen.bmj.com/content/10/1...
Probably because it is a big effort to check the code? Biometrical Journal does it. They have reproducibility editors who do excellent job. As an AE or reviewer for any journal, I always require code to be available.
March 13, 2025 at 8:05 PM
Probably because it is a big effort to check the code? Biometrical Journal does it. They have reproducibility editors who do excellent job. As an AE or reviewer for any journal, I always require code to be available.
SPOILER: Forget "naive" confidence intervals computed by considering the results in the K folds as iid observations...
The K folds in K-fold-cross-validation are NOT independent!
And this cannot just be ignored...
The K folds in K-fold-cross-validation are NOT independent!
And this cannot just be ignored...
March 5, 2025 at 8:00 AM
SPOILER: Forget "naive" confidence intervals computed by considering the results in the K folds as iid observations...
The K folds in K-fold-cross-validation are NOT independent!
And this cannot just be ignored...
The K folds in K-fold-cross-validation are NOT independent!
And this cannot just be ignored...
Reposted by Anne-Laure Boulesteix
IMO it’s a mistake to give stats to 1st year med students. It’s not why they chose medicine & they resent it. Better to wait until they have developed some curiosity for it. I argued unsuccessfully for this during my time at UCL.
November 25, 2024 at 7:49 AM
IMO it’s a mistake to give stats to 1st year med students. It’s not why they chose medicine & they resent it. Better to wait until they have developed some curiosity for it. I argued unsuccessfully for this during my time at UCL.
I could nnot agree more. One problem is that for some packages (not those you are involved in ;-)) it is difficult to figure out which method/theory they implement. The link between theory and implementation is not well documented. So people just mention the package name in place of the method...
November 19, 2024 at 12:01 PM
I could nnot agree more. One problem is that for some packages (not those you are involved in ;-)) it is difficult to figure out which method/theory they implement. The link between theory and implementation is not well documented. So people just mention the package name in place of the method...
Reposted by Anne-Laure Boulesteix
AI researchers: hold my beer —
Old model performed slightly not worse when data were generated by the new generative AI method I just made up
Old model performed slightly not worse when data were generated by the new generative AI method I just made up
November 18, 2024 at 9:19 AM
AI researchers: hold my beer —
Old model performed slightly not worse when data were generated by the new generative AI method I just made up
Old model performed slightly not worse when data were generated by the new generative AI method I just made up
Reposted by Anne-Laure Boulesteix
Which reminds me of this Andrew Vickers quote
www.nature.com/articles/ncp...
www.nature.com/articles/ncp...
November 11, 2024 at 9:53 AM
Which reminds me of this Andrew Vickers quote
www.nature.com/articles/ncp...
www.nature.com/articles/ncp...