Experimental design for causal inference through an optimization lens

J Zhao - … in Operations Research: Smarter Decisions for a …, 2024 - pubsonline.informs.org
The study of experimental design offers tremendous benefits for answering causal questions
across a wide range of applications, including agricultural experiments, clinical trials …

Beyond Neyman–Pearson: E-values enable hypothesis testing with a data-driven alpha

PD Grünwald - Proceedings of the National Academy of Sciences, 2024 - pnas.org
A standard practice in statistical hypothesis testing is to mention the P-value alongside the
accept/reject decision. We show the advantages of mentioning an e-value instead. With P …

Anytime-Valid Continuous-Time Confidence Processes for Inhomogeneous Poisson Processes

M Lindon, N Kallus - arXiv preprint arXiv:2410.09282, 2024 - arxiv.org
Motivated by monitoring the arrival of incoming adverse events such as customer support
calls or crash reports from users exposed to an experimental product change, we consider …

YEAST: Yet Another Sequential Test

A Kurennoy, M Dodin, T Gurbanov… - arXiv preprint arXiv …, 2024 - arxiv.org
Large-scale randomised experiments have become a standard tool for developing products
and improving user experience. To reduce losses from shipping harmful changes …

Basemath's Test: Group Sequential Testing Without Recurrent Numerical Integration

A Weiss - Available at SSRN 4599695, 2024 - papers.ssrn.com
Basemath's Test is a novel (Group) Sequential Testing approach. This method simplifies
statistical analysis by eliminating the need for recurrent numerical integration. The presented …