Profile avatar
thomas-speidel.bsky.social
31 posts 74 followers 524 following
Regular Contributor
Active Commenter
Posts 13 Comments 31

@cderv.bsky.social @cscheid.net Fantastic work on _brand.yml ! Does it work with Quarto books? I'm playing with it and it seems to be ignored.

submitted 141 days ago • 1 comment

NEW PAPER Really glad to see this one in print: the harm due to class imbalance corrections in prediction models developed using ML/AI Excellently led by @alcarriero.bsky.social onlinelibrary.wiley.com/doi/epdf/10....

submitted 148 days ago • 0 comments

What large language models know and what people think they know https://www.nature.com/articles/s42256-024-00976-7 🧬🖥️🧪 https://osf.io/y7pr6/

submitted 153 days ago • 1 comment

NEW PREPRINT A detailed overview of 32 popular predictive performance metrics for prediction models arxiv.org/abs/2412.10288

submitted 190 days ago • 10 comments

Hi, can you help me? I want to develop a model that makes risk predictions. Use logistic regression. Can I use some more modern techniques, like AI? Use a neural network with single non-hidden feed forward layer that outputs to a single dimension using a sigmoid activation function.

submitted 215 days ago • 19 comments

I've begun a statistical graphics starter pack at go.bsky.app/EMb8xWf . Please help me find more people or feeds to add to this pack. #rstats #visualisation #statistics 📈

submitted 217 days ago • 26 comments

Great to see this R package: stephenrho.github.io/pminternal/ Implements the (in)stability plots and checks recommended by myself & @gscollins.bsky.social after developing a prediction model Our paper is here: onlinelibrary.wiley.com/doi/full/10....

submitted 217 days ago • 1 comment

Characterising #machinelearning studies in healthcare, the full set Weak design/methods tinyurl.com/yc4easr9 Poor reporting tinyurl.com/55ed3j9k High risk of bias tinyurl.com/yk6m9sx5 Full of Spin tinyurl.com/yckubrnp Not open science tinyurl.com/437bfz8f #StatsSky #MLSky #mustdobetter 😬

submitted 223 days ago • 8 comments

New statistical significance meme! #StatsSky #EpiSky

submitted 222 days ago • 4 comments

New blog post! I helped build Idaho's official election results website (with #rstats and #quartopub) and we did some neat stuff to auto-generate hundreds of R chunks. Here's a guide about how to create tabsets and slides programmatically with inline knitr chunks www.andrewheiss.com/blog/2024/11...

submitted 232 days ago • 10 comments

#statistics thought of the day: In the absence of detailed knowledge (e.g. of measurement errors) about a numeric variable, a useful descriptive statistic is the variable's relative information content: stats.stackexchange.com/questions/65...

submitted 250 days ago • 1 comment

Gang, what are the most common problems you see around p-values in the social science literature—misinterpretations, misuse, misunderstandings, mis...p-valuing?! Thanks 🙏

submitted 251 days ago • 21 comments

Quanta magazine story about multiple imputation #stats www.quantamagazine.org/when-data-is...

submitted 251 days ago • 0 comments