Sebastian Hellmann
PostDoc working at TU Munich.
Interested in on computational modelling, decision-making, and confidence.
Cat owner, Ireland lover and brass music fan
- For all who use Bayesian hierarchical models, have a look at our new preprint, out now together with @linushof.bsky.social @nunobusch.bsky.social and @thorstenpachur.bsky.social osf.io/preprints/ps...
-
View full threadThanks for the kind words. Glad to see that people already account for this in some packages. Not sure whether it helps a lot, but using the direct computations instead of using numerical integration may still speed up things a bit (about 20 times on my machine)
- Unfortunately, for the variance on the probability scale, the speed up vanishes.
- (e.g. the standard normal CDF) and normal distributions to fit the group-level distribution. But we cannot simply apply the same transformation to the mean of the real-valued normal distribution to derive the group-level mean on the parameter scale! This ignores individual variability.
- The good news: We provide a simple, correct computation that accounts for this variability, ensuring accurate group-level inferences. This fix is crucial for reliable conclusions in all cognitive models with constrained parameters. Check out the details to improve your hierarchical analyses!
- If you’re estimating group-level means of constraint parameters, which are fitted with nonlinear transformations, beware that a common approach can produce biased estimates—especially with high individual variability. For constraint parameters, we often use nonlinear transformations...
- Thanks so much, also for your support and your wisdom. It has been a pleasure to be your Padawan ;)
- Massive congrats @sehellmann.bsky.social for his well-deserved win of the Yearly Price for the best dissertation at the KU!
- [Not loaded yet]
- Hey, could you add me, too, please? Thanks a lot!
- Thanks for teaching me so much @manurausch.bsky.social !
- New preprint together with @manurausch.bsky.social about the importance to consider decision time when reflecting on the accuracy of a decision, showing that considering decision time in confidence is Bayes optimal and human observers seem to make use of this information. osf.io/preprints/ps...
