New missing data paper just dropped, in collaboration with my friends @ABarPhD and @DrStefanyC! =D
https://t.co/sEwoF4LWB9
Come for the data mining, stay for a discussion of some oft-overlooked nuances of different MAR missing data mechanisms. 1/
Just in time for conference season: Our paper on simulation studies in psychology was accepted at Psychological Methods! 🥳See the thread below for a summary, the updated preprint is available at https://t.co/Re3krDCv1F
My first day @CityofEdmonton@UAlberta - here to create a Centre for Open Science & Synthesis in #Ecology & #Evolution. Looking for ~5 postdocs to join me in establishing the centre. Please apply - https://t.co/iwdRzcEaur Please RT
@SolomonKurz@rabaath Really like this type of layout but would also prefer it rotated 90° so the trace plots are horizontal (which is how I did it in "mitml": https://t.co/eYXDXYUssA)
@PWGTennant The "intellectual theft machines" is more drastic than I would put it, but you're summarizing some of my own feelings as well. Most people I talk to cite "gotta go fast" as a motivation to use LLMs, and I think this is as much an argument against them as it is in their favor.
New preprint 📝 Multiple imputation of missing data in large studies with many variables: A fully conditional specification approach using partial least squares.
Preprint: https://t.co/zcQEghTKyi
OSF-Repo: https://t.co/cTs6xsyG5d
Our aim was to describe/evaluate an extension of the FCS ("chained equations", MICE) algorithm for MI that incorporates dimension reduction steps for large studies with many variables.
Results suggest that this way of specifying imputation models could be very flexible and ...
My third and final dissertation paper was published at EJP!✨ We tested how #neuroticism and romantic experiences in #adolescence jointly contribute to differences in affect level and variability. Check out the #openacess article: https://t.co/8uVIg3Cvvc
We reviewed 100 psych. simulation studies & find room for improvement in planning/reporting. As a remedy, we (@BartosFra, @tmorris_mrc, @BoulesteixLaure, @Daniel_W_Heck & Samuel Pawel) present ADEMP-PreReg, a sim study preregistration & reporting template
https://t.co/kmWdgydo5F
@jd_wilko Not sure what tool this is, but you can get the same effect by other means (e.g., if the tool generates floathing point or integer numbers in a large-enough range, then you can copy/paste them in Excel next to a list of 1s and 2s and sort).
@jd_wilko That being said, there *is* something to say about stochastic group sizes being an underacknowledged issue in experiments/RCTs: https://t.co/YHa8Yp4wvv
@jd_wilko I always understood it to mean that most people just randomize a pre-defined list? Most software can do this just fine (sample(rep(1:2, each = 100)) in R), and online generators do it too (https://t.co/ih1StGSToF).
Niche question: Any of my tweeps know any Editors/IDEs for Windows that have both #Rstats supports and a decent Vim mode?
Getting increasingly frustrated with RStudio, looked at VSCode but found it hard to configure Vim-like keybindings.