New post based on an interesting recent conversation âTest-based confidence intervals or confidence-interval-based tests?â open.substack.com/pub/tpmorris...
Confidence intervals and tests often correspond. That is, if a one-sided test returns p=0.05 then the bound of a one-sided confidence interval should touch the null H0; if a one-sided test returns p>0...
The Metascience conference is coming to London! June 30-July 2, 2025
I was also surprised that it is not open access... Hopefully an error that will be quickly be corrected.
NEW PAPER: Data are often imperfect, do not fulfill assumptions of methods, etc. What is the impact on the reliability of results? Abrahamowicz et al. (with @timpmorris.bsky.socialacademic.oup.com/aje/advance-...
Abstract. Quantitative bias analysis (QBA) permits assessment of the expected impact of various imperfections of the available data on the results and conc
JAMA introduces a new framework for drawing causal inferences from observational data. Including a call for adopting explicit causal language. This is a key moment in the embrace of contemporary causal inference methods within health science! jamanetwork.com/journals/jam...#EpiSky#CausalSky
This study examines drawing causal inferences about the effects of interventions from observational studies in medical journals.
Theyâre trying to get a hold on the jungle of cluster analysis. statmodeling.stat.columbia.edu/2024/05/12/t...
An update on Gino's lawsuit against us (Data Colada). We had a hearing about our motion to dismiss, and we've learned some legal things along the way. No big news yet, but some details: datacolada.org/116
âAny update on the lawsuit?â That is the most common question any of us is asked. It is usually preceded by an apologetic preamble, like, âsorry if this is a sensitive question,â or âI donât know if youâre tired of talking about this, butâŚâ The reality is that, for the most part, our actual sensitivity...
I'm excited at the prospect of working closely with someone on the implementation of what promises to be an impactful project! Don't hesitate to ask me about the role!
co-authors are @iamjulianlange.bsky.social K. Eggensperger, G. Casalicchio, M. Wever, M. Feurer, D. RĂźgamer, E. HĂźllermeier, myself and B. Bischl
NEW accepted ICML'24 position paper: âRethinking Empirical Research in Machine Learning: Addressing Epistemic and Methodological Challenges of Experimentationâ by Moritz Herrmann et al.: arxiv.org/abs/2405.02200
We warn against a common but incomplete understanding of empirical research in machine learning (ML) that leads to non-replicable results, makes findings unreliable, and threatens to undermine...