Statistician at the University Medical Center Göttingen (UMG) Views are my own. medstat.umg.eu
🤔 Should one meta-analyze a single study?
☝️🤓 It turns out this makes sense in certain cases - if you're looking for predictions ("MAP priors"). In the particular case of a single study, there are also close links to power priors and bias allowance models. Have a look:
➡️ https://arxiv.org/abs/2505.15502
(joint work with @friede1 )
For people interested in methods for #MetaAnalysis
Symposium: Recent Advances in Meta-Analysis
Dortmund University, Germany, June 30th - July 1st, 2025
https://msind.statistik.tu-dortmund.de/en/research/events/meta-analysis-symposium-2025/
The Society for Research Synthesis Methodology (SRSM)'s annual meeting will take place in Madison, Wisconsin this year.
See here for more details:
➡️ https://www.srsm.org/meetings.html
Save the date! useR! 2025 is coming in Aug 8-10, 2025! Use the QR code to be notified when registration opens!
"Statistics in times of AI"
Short reminder that there are only five days left to submit an abstract to the DAGStat 2025 conference.
The EU-funded 🇪🇺 INVENTS project (https://invents-he.eu) held its first annual consortium meeting this week in Göttingen. Looking forward to further fruitful collaboration!
A kick-off meeting of researchers from Japan and Germany within the #HeKKSaGOn network (www.HeKKSaGOn.net) came to a successful finish today in Göttingen. Looking forward to further collaborations!
15 years ago the #DFG opened an office in #Tokyo in response to the strong interest in #researchcooperation with partners in #Japan. Since then our office evolved into an essential point of contact for the research communities. Happy Anniversary to our colleagues and all partners!🎉
Clinical trials are usually designed to demonstrate an overall treatment effect, but are often not able to resolve effects in patient subgroups. We performed a #MetaAnalysis, considering several studies to look for differences in treatment effects between patient subgroups in #MultipleSclerosis and investigate which patients may benefit most (or least) from immunotherapy. See here for more details:
➡️ https://doi.org/10.1177/20552173241274618
(2) Wir haben aber schon ein bisschen vorgefeiert, und zwar letzte Woche in der Vertretung des Landes Baden-Württemberg in Berlin: mit einem parlamentarischen Abend.
Neben Vorträgen des Patientenbeauftragten der Bundesregierung Stefan Schwartze bzw. seiner Referentin Theresa Berhorn und des Vorsitzenden von @CochraneD, Jörg Meerpohl, gab er viel Gelegenheit zum Austausch inmitten einer tollen Ausstellung von Axl Jansen.
Hier ein paar Eindrücke.
Lange nichts mehr gepostet hier, sorry! Wir haben immer noch nicht den richtigen Dreh gefunden, da große Teile unserer ehemaligen Zielgruppen sich leider immer noch bei X herumtreiben, zu Bluesky oder Threads gegangen sind - oder die sozialen Medien gleich ganz verlassen haben. Aber diese Woche haben wir ein paar spannende News:
(1) Das IQWiG-Herbst-Symposium am 29.11. dreht sich um "unterhaltsame #Wissenschaftskommunikation". Man kann sich schon anmelden: https://www.iqwig.de/veranstaltungen/herbst-symposium/
My book "Bayesian Statistics: The Basics" is almost finished! If you would like to take a look at an early version and possibly provide a brief review, let me know.
In exchange for your review, you'll get a free copy of the book when it is published (and my sincere appreciation!) :)
All three books I've co-authored are freely available online for non-commercial use:
#Bayesian Data Analysis, 3rd ed (aka BDA3) at https://stat.columbia.edu/~gelman/book/ and lectures plus #rstats, #Python and #Stan code at https://avehtari.github.io/BDA_course_Aalto/
#Regression and Other Stories at https://avehtari.github.io/ROS-Examples/ including #rstats and #Stan code
Active Statistics at https://avehtari.github.io/ActiveStatistics/
The websites also have links to the publishers' web stores if you prefer hard copies of these
⚽ Probabilistic forecast for today's #EURO2024 final: #ESPENG
🏆 Probability to win: 47.0% #ESP vs. 53.0% #ENG
🥅 In normal time:
Mean goals: 1.2-1.3
🇪🇸 33.9%
Draw 26.8%
🏴 39.2%
Heatmap with probabilistic forecasts for the possible outcomes of the match in normal time:
We're hosting a symposium on
"Recent advances in #MetaAnalysis"
on August 28/29 in Göttingen.
For more details and the line-up of speakers, see here:
➡️ https://medstat.umg.eu/aktuelles/symposium-meta-analysis-2024/
and stay tuned for updates!
#MedStatGoe
Causal Inference in R
https://www.r-causal.org/
The book is in the making. What is available so far looks useful for teaching and accessibly written #rstats #CausalInference
PSA: All #rstats package on #cran will get an official DOI!
This will facilitate bibliometrics and giving credit to :rstats: package authors.
Registering all 20,000+ packages will still take a few more days. But the first couple of thousand are already live. Example:
(2/2) The book is open and available online 👇🏼
https://www.r-causal.org
Thanks to the authors for making this resource available for free! 🙏🏼
Image credit: Book website
(1/2) Causal Inference in R 🚀
If you are doing or want to learn causal inference, I recommend you check Causal Inference in R - a new book by Malcolm Barrett, Lucy D’Agostino McGowan, and Travis Gerke. The book focuses on the foundation of causal inference, and covers the following topics:
✅ Setting causal questions
✅ Setting assumptions and identifying the target population
✅ Experiment design
✅ Fit causal models and estimate the causal effects
✅ Sensitivity analyses
#rstats R version 4.4.1 "Race for Your Life" (source version) has been released.
Dedicated to the memory of Fritz Leisch. He was fond of riding his motorcycle, which is reflected in the nickname.
The binaries for Macs will be delayed by a couple of days because Simon Urbanek is traveling.
