PhD at the @ScanUnit 🧠, University of Vienna 🎡🇦🇹. Passionate about Stats and Modelling 🧮🎲. Triple dog dad 🐶🐶🐶
Vienna
Joined February 2013
- Tweets257
- Following166
- Followers164
- Likes225
Pinned Tweet
Happy to finally launch my blog! I plan to post every few weeks about statistical topics I find interesting.
I start with a few thoughts on the question: are significant results from low-powered tests less credible?
lukelikelihood.com/2020/05/1…
Anyone willing to give the health experts in Germany (and Denmark, and Norway, and Ireland, and others) a crash course in the base-rate fallacy?
Lukas Lengersdorff retweeted
Words I've found useful to explicitly ban in discussion with more junior researchers:
- dumb
- stupid
- trivial
- obvious
I try not to use them and I expect them not to use them either. What else should go on the list?
Lukas Lengersdorff retweeted
As a statistician, there’s a big secret that I think the public needs to know. Statisticians don’t all agree on what probability is! Yes we use it in all our work but there’s been many brutal debates and we’re still not 100% sure what it is so we mostly just agree to disagree.
Lukas Lengersdorff retweeted
Another successful pub quiz, this time hosted by @LukeLikelihood and with @lei_zhang_lz as the glorious winner! 🥳@GuranCNA
#AcademicTwitter, now it's your turn! Can you guess all the book titles/authors that got translated badly and find the correct names of the theorems? ↙️
Live footage of people all over the world reacting to the end of Donald Trump's presidency:
youtube.com/watch?v=GlCFPo6Y…
Am I the only one who‘s trypophobia is slightly triggered by this graph? 🙃
You’re unable to view this Post because this account owner limits who can view their Posts. Learn more
Today I realized that I really love #Vienna, and the feeling of safety and comfort the city and its people provide. The peace and freedom we are allowed to enjoy in Vienna (and Austria) are really a freaking precious miracle we should be thankful for (and care for) every day.
Lukas Lengersdorff retweeted
Passt auf euch auf!
Bleibt daheim!
Verteilt keine Falschinformationen!
#wienliebe
Please stay safe, everybody! 😢
CONFIRMED at the moment:
*08:00 pm: several shots fired, beginning at Seitenstettengasse
*several suspects armed with rifles
*six different shooting locations
* one deceaced person, several injured (1 officer included)
*1 suspect shot and killed by police officers #0211w
Lukas Lengersdorff retweeted
Wenn Sie Videos oder Fotos des derzeitigen Einsatzes im 1. Bezirk haben - HIER uploaden, NICHT in sozialen Medien verbreiten! So können Sie uns unterstützen! Bitte um zahlreichen RT
upload.bmi.gv.at/ #0211w
Lukas Lengersdorff retweeted
SIPS Commendation winner BayesCog lecture series is an introductory course on Bayesian statistics and hierarchical Bayesian modeling using Stan, dedicated to the psychology & cognitive (neuro-)science audience. teapawt.net/lei_zhang_lz/status/12…
1/13) This semester's teaching on Bayesian stats and cognitive modeling is over! Thanks to COVID (ironically!), I recorded all my teaching sessions w/ @zoom_us, and they are available on #Youtube.
Wondered what have we covered to the cog-neuro audience? A thread.
Lukas Lengersdorff retweeted
I don’t know who needs to hear this but people who happily analyze data without worrying about philosophy are usually unthinkingly following the ideas of some dead philosopher.
Lukas Lengersdorff retweeted
OK, friends, this is a big one. How Eugenics Shaped Statistics, my cover story for this month’s issue of Nautilus, is now online.
I have just found a new life goal:
I will organize a statistical symposium on the works of George Box, with the buffet exclusively provided by George's Box: cafegeorge.at/georges-box/
Lukas Lengersdorff retweeted
The ultimate summary on #rstats vs. #machinelearning.
Anybody knows where this is from?
@kareem_carr @daniela_witten @MaartenvSmeden
I highly recommend watching the recording of the recent Statistics Debate: youtube.com/watch?v=_D8U58QL…
Example questions: Should we abandon p-values? Should we use Bayes Factors?
With Deborah Mayo, Jim Berger, & David Trafimov.
Organized by @NISS_DataSci
It is comforting to hear someone say that it really IS important to consider that freq and Bayes answer very different questions. Lately, I often heard the argument that it is not important that the two methods answer different questions, as long as they give the same answer.
Here's an interesting "live" thread on the debate: teapawt.net/lakens/status/13167736…
Listening to the #StatisticsDebate. niss.org/events/statistics-d… Strong opening by @learnfromerror that p-values should be used. Now Jim Berger is talking about how they are misunderstood. True, but there is no evidence any alternative method will do better.