ERCIM News No. 145 Special theme: "E-Values - Statistical Testing for the 21st Century"

ERCIM NEWS No. 145 -

Dear ERCIM News reader, ever heard of e-values?

If not, you're probably not alone. 

Michael I. Jordan opens this issue with a keynote introducing the concept:

What is an “e-value” and why has it become an object of intense study in statistics and in the allied fields of machine learning, signal processing, and econometrics? To briefly introduce the basic idea, let us consider one of the core problems in statistics – the “hypothesis testing problem” of deciding whether observed data is consistent with some particular data-generating mechanism (often referred to as a “null hypothesis”) or is better explained by another mechanism (referred to as an “alternative hypothesis”). This problem is addressed by defining some function of the data (a “statistic”) whose distribution is as different as possible under the null and the alternative. Given an observed value of such a statistic, one then makes a choice between the two distributions, doing so in a way that minimizes the probability of errors. Classical statistical theory provides a unifying framework – the “p-value” – by which the choice between the null and alternative hypotheses reduces to a thresholding procedure.

Guest editors Peter Grünwald (CWI and Leiden University), Wouter Koolen (CWI and University of Twente) and Johanna Ziegel (ETH Zurich) guide readers through this emerging framework for quantifying statistical evidence. 

The articles in this special theme demonstrate how it is already finding applications across many areas of scientific research, from clinical trials and microbiome studies to political polls, AI evaluation and financial risk backtesting.

 

<>

I must admit when I read 'e-values', I thought of an e-rendering of ethics - humanities and sciences ( qualitative-quantitative : subjective-objective : ... ), but as always a welcome resource.

Includes:

p.6 Introduction to the Special Theme - E-values Statistical Testing for the 21st Century
by the guest editors Peter Grünwald (CWI and Leiden University), Wouter Koolen (CWI and University of Twente) and Johanna Ziegel (ETH Zurich) 

p.8 On Testing by Imaginary Betting^
by Glenn Shafer

p.37 Large Language Models as Design Partners: Automating Graphical Mockups to Refine Requirements
by Giovanna Broccia, Maurice H. ter Beek (CNR-ISTI), and Alessio Ferrari (University College Dublin and CNR-ISTI) 

p.39 Challenges in Small-Scale Medical Data Exchange Platform Development
by Hubert Schölnast, Peter Kieseberg, Patrick Kochberger and Henri Ruotsalainen (University of Applied Sciences St. Pölten) 

p.42 SCIANCE: AI for Scientific Discovery in Europe
by András Benczúr, Edina Nemeth (SZTAKI), Jonas L'Haridon (European Science Foundation) and Magdalena Brus (EGI Foundation)

My source: Peter Kunz c/o ERCIM News