Tipo de evento: charla
Organiza: Data Observatory/ Universidad Adolfo Ibáñez
Horario: 15:00 horas
Tema: Kernelized log-rank tests for Survival Analysis data
Expositora: Tamara Fernández, UAI
Organiza: FIC UAI/ DO
Weighted log-rank tests are arguably the most widely used tests by practitioners in the context of Survival Analysis data. Despite their popularity, there are many instances in which their performance is far from optimal, and thus many approaches have been considered to make them more robust against a broader family of alternatives, including taking linear combinations, or the maximum among a finite collection of them. In this talk we introduce a general non-parametric test for right-censored data which is based on the «kernelization» of weighted log-rank tests. We show our test-statistic has a dual interpretation, first in terms of the supremum of a potentially infinite collection of weighted log-rank tests, with weight functions belonging to a Reproducing kernel Hilbert space of functions (RKHS), and second, as the norm of the difference of the embeddings of certain «depencency» measures into the RKHS. We provide an easy-to-use test-statistic as well as an economic Wild-Bootstrap procedure and study asymptotic properties of the test, finding sufficient conditions to ensure our test is omnibus. We show our testing procedure performs, in general, better than competing approaches.
* Tamara Fernández obtained her Dphil in Statistics in 2018 from the University of Oxford. After her Dphil, she received a Biometrika Postdoctoral Fellowship at the Gatsby Computational Neuroscience Unit at University College London. From October 2020, she is an Assistant professor at the Universidad Adolfo Ibáñez in Chile. Her research interests are mostly focused on the development of statistical methodology/theory based on non-parametric methods combined with Machine Learning techniques. Her work has been mostly applied to Survival Analysis data.
Enlace zoom: https://zoom.us/j/98116049572