Posts by Štěpán Jabůrek
Demo. Of course, I created it with major (claude code) vibecoding. Try it out and never be a bored Bayesian again! Or you can just open the animations by Chi Feng manually 😀 - chi-feng.github.io/mcmc-demo/ap...
The idea came to me yesterday on the subway. Having to wait several minutes(!) for a model to fit is insane (to a recovering frequentist). But I loved the sampling animations of MCMC and thought it would be fun to look at them while running the model.
Bored while your Bayesian model samples? Try my basically useless (and massively vibecoded) R package "bayeswatch🏝️"! You can watch some nice MCMC animations directly in your RStudio/Positron while you wait!
"bayeswatch" simply wraps animations by Chi Feng on top of brms and rethinking 🧵
I mainly want to thank my future dissertation supervisor Michal Parizek and my current diploma supervisor Jakub Tesař for their continued support and help. I am really happy to continue at my alma mater and looking forward to this new challenge🔽
🥳Extremely happy that I have been accepted for a PhD position in International Relations at @fsv.charlesuni.cuni.cz. I will (hopefully) research the politicization of international economic cooperation in Europe using NLP and quantitative methods in a IR + party politics setting 🔽
While the general trends are comparable, the Social Dems (ČSSD) and Civic Dems (ODS) show relatively bigger swings in discoursive stances and the last years show a swap of overall stances (my prior is that the Ukraine war drives this). New CHES should drop this year. 2/
Had an amazing time at my first #COMPTEXT2025 in Vienna. Happy to see so much cool research, get the chance present our paper and represent a growing team of researchers from the Institute of Political Studies @fsv.charlesuni.cuni.cz
A visualisation of the Gulf of México as illuminated contours
Illuminated contours of the Gulf of México.
#rayshader adventures, an #rstats tale
The fires are revealing how good elected officials are at their jobs.
Unbiased information systematically affects electoral fortunes.
This is why natural disasters being bad for incumbents on average isn't evidence of voter irrationality.
voices.uchicago.edu/ethanbdm/fil...
Bee-nado, meet sharknado! :P 🦈
#rayvertex adventures, an #rstats tale