New e-value even more flexible: significance level adjustable at a later stage
Recently, a paper by Peter Grünwald was published in the prestigious scientific journal Proceedings of the National Academy of Sciences. Grünwald is full professor of Statistical Learning at the Mathematical Institute and senior researcher at Centrum Wiskunde & Informatica.
It was already known that e-values are more flexible than p-values: with e-values you can stop an experiment earlier than originally planned or, for example, add subjects afterwards. In this paper, entitled Beyond Neyman–Pearson: E-values enable hypothesis testing with a data-driven alpha, Grünwald shows that e-values are also more flexible in another way: with e-values, it is possible to determine the significance level at a later time than usual.
Read the entire article on the website of CWI.
About Peter Grünwald
Peter Grünwald is appointed as parttime full professor of Statistical Learning at the Mathematical Institute of Leiden University and senior researcher in the Machine Learning group of Centrum Wiskunde & Informatica (CWI). He has previously received fundings, such as a VIDI and VICI via NWO and an ERC Advanced Grant, and in 2010 he was co-recipient of the Van Dantzig prize, the highest Dutch award in statistics and operations research. He has a great interest in foundations of statistics and regularly gives talks about the problems and difficulties surrounding traditional statistical methods.
Text: with thanks to Marije Huiskes-Tolsma (CWI)
Photo: CWI