stippe87.bsky.social - Profile | ThreadSky | a Reddit-style client for Bluesky

I am approaching a new field, and I realized once again that one of the main issues with code written by scientists who are not familiar with software development is reproducibility, making those ideas much harder to spread

submitted 5 days ago • 0 comments

New post: how to work with multiple (2+) level models in Bambi. thestippe.github.io/statistics/t... #stats #python #pymc #bambi

submitted 10 days ago • 0 comments

Ieri c'è stata la prima lezione, davvero interessante, di introduzione alla filosofia della scienza, con una spiegazione delle principali correnti di pensiero. Grande affluenza e lezione bellissima. Da stasera si entra nel cuore della discussione.

submitted 10 days ago • 0 comments

I just realized a lot of ppl totally misunderstood the purpose of unsupervised models: they think that you need it when you have unclear questions. I will tell you a secret: there's no model, no AI, nor any other tool which can help you if you have unclear questions. #stats

submitted 11 days ago • 0 comments

An infinite particle physics debate seems closes! It looks like the muon g-2 puzzle has been solved! www.science.org/content/arti...

submitted 11 days ago • 0 comments

@davidkohns.bsky.social has extended our paper "The ARR2 prior: flexible predictive prior definition for Bayesian auto-regressions" doi.org/10.1214/25-B... with a case study showing how to use ARR2 prior with dynamic (time-varying parameter) regression models and Stan

submitted 12 days ago • 0 comments

🤣🤣

submitted 18 days ago • 757 comments

A large section of a glacier in the Swiss Alps has broken off, resulting in a torrent of ice, mud, and rock that has submerged the majority of a village evacuated earlier this month as a security precaution. #EuropeNews

submitted 18 days ago • 1 comment

Omg, 😍. “This is a site for pictures of owls… in towels.” owlsintowels.org

submitted 19 days ago • 160 comments

In the morning I gave a talk about Bayesian cross-validation at KU Leuven and in the afternoon got to wear Belgian academic gown and hear David Spiegelhalter's honorate doctorate talk, which was great

submitted 19 days ago • 2 comments

Microsoft fires thousands of employees across different countries while investing a lot in AI. The companies' services are getting worst and worst, nothing works as it should. But companies still rely on Microsoft. Why? www.cnbc.com/2025/05/13/m...

submitted 19 days ago • 1 comment

New post on hierarchical models with Python. How to use Bambi and PyMC to write models in a lme4/brms-like way. thestippe.github.io/statistics/b... #stats #python

submitted 19 days ago • 0 comments

💈 La disinformazione e la propaganda del governo sul referendum cittadinanza | @vitalbaa.bsky.social www.valigiablu.it/referendum-c...

submitted 20 days ago • 3 comments

I should do a bot named "Idiot posts from linkedin". Today's one would be "using adjectives slows down productivity".

submitted 20 days ago • 0 comments

ANADOLU AGENCY: Photos of the week www.anadoluimages.com/Search?colle... @anadoluajansicom.bsky.social #photojournalism #journalism #photography

submitted 21 days ago • 0 comments

Another great episode of Normal Curves www.normalcurves.com/sugar-sag-is... - the phenomenon of "references to nowhere" resonated with me - following a trail of references which never get to a source - and in this case involved plagiarism. #statistics #methodology

submitted 27 days ago • 0 comments

xkcd on radon statmodeling.stat.columbia.edu/2025/05/24/x...

submitted 23 days ago • 1 comment

One of the worst myths of data science is that you need tons of data to fit good models. All you need is a good sample, quantity is not a valid replacement for quality.

submitted 21 days ago • 0 comments

Le immagini che Trump ha presentato come prove del "genocidio dei bianchi" in Sudafrica sono state prese da un blog conservatore e non c’entrano nulla con il Sudafrica: ilpost.link/2vVVxi2GAA

submitted 24 days ago • 2 comments

🧵 “La nuova incursione militare a Gaza è moralmente ingiustificabile”. Il Regno Unito sospende i rapporti commerciali con Israele www.valigiablu.it/regno-unito-...

submitted 26 days ago • 3 comments

My brothers and sisters who want to send a reporter a tip: please use ONE font size and ONE font and ONE font color. Otherwise it feels like the email equivalent of a ransom note made of cut-out letters

submitted 27 days ago • 1 comment

FdI, partito di un presidente del Senato che invita a non andare a votare per un referendum, critica il gesto di non voler indossare il tricolore. Mi pare tanto il bue che da del cornuto all'asino.

submitted 26 days ago • 0 comments

Chi, come #Fedriga, voleva restare al potere oltre i limiti in #FVG, oggi è smentito dalla #CorteCostituzionale: il #terzomandato è incostituzionale. Le Regioni a Statuto Speciale non sono zone franche dai principi democratici. Serve #ricambio e #trasparenza www.facebook.com/photo/?fbid=...

submitted 32 days ago • 0 comments

If only Arxiv required researchers to sign at the top rather than the bottom of the page, none of this would’ve happened. statmodeling.stat.columbia.edu/2025/05/19/i...

submitted 28 days ago • 0 comments

The interesting question is why so many pundits and reporters — and, it seems, small stock investors — have been sounding the all clear on Trump’s tariffs, when the reality is that all we’ve seen is a modest retreat from complete, destructive insanity to seriously harmful madness.

submitted 31 days ago • 0 comments

A massive health dataset has spawned a wealth of cookie-cutter "research Mad Libs" papers that don't tell us anything useful, but flood the literature with noise. Possibly AI-generated, possibly paper mill origin. www.science.org/content/arti...

submitted 32 days ago • 2 comments

submitted 33 days ago • 11 comments