Seminar Series on the Mathematics of Data Science - Department of Applied Mathematics
With the MDS Seminar, we would like to launch a lecture series in which both researchers from the University of Twente and external researchers present their current work in the field of mathematics of data science. The aim is to get to know and understand the research of other groups and disciplines better. It offers the opportunity for regular exchange as well as a basis for possible collaborations.
Format
Seminars are held on campus and via Teams. All seminars occur every fortnight on Mondays at 4 p.m. unless otherwise stated (see the program below for the dates and the rooms).
Upcoming seminars
02 February 2026, 11:00 (CR 2M)
- Speaker: Richard Post (Erasmus MC)
Title: The complex causal interpretation of common effect measures in time-to-event analysis
Abstract:
n recent years, the hazard ratio's causal interpretation (or lack thereof) has gained significant attention. In the presence of unobserved heterogeneity, even for data from a randomized controlled trial, the observed hazard ratio (OHR) suffers from built-in selection bias as, over time, the individuals at risk in the exposed and unexposed are no longer exchangeable. In this talk, I will examine a general structural causal model (SCM) to formalize how the observed hazard rate evolves. When the causal effect of an exposure on the hazard is multiplicative, the expected OHR can be shown to equal the ratio of expectations of the latent variables (frailty and modifier) conditioned on survival with and without exposure. Thus, in the presence of frailty and/or unobserved effect modifiers, the expected OHR deviates from the population effect of interest. Moreover, if the causal effect is additive, the expected observed hazard difference (OHD) evolves by selecting favorable levels of effect modifiers in the exposed group. Consequently, in the presence of unobserved effect heterogeneity, the OHD deviates from the population causal effect of interest. Finally, I will explain why acceleration factors and contrasts of survival functions offer clear causal interpretations and are thus more suitable for describing causal effects. Finally, I will formalize the causal interpretation of the acceleration factor and show that it is a valid causal effect measure, even in the presence of frailty and treatment-effect heterogeneity.
(Based on https://doi.org/10.1007/s10985-024-09616-z, https://doi.org/10.1007/s10985-024-09617-y and https://arxiv.org/abs/2409.01983)
16 February 2026, 16:00 (RA 2334)
- Speaker: Fabian Mies (TU Delft)
Title: T.b.a.
02 March 2026, 11:00 (CR 2L)
- Speaker: Ivo Stoepker (TU/e)
Title: T.b.a.
16 MaRch 2026, 16:00 (RA 2504)
- Speaker: dr. ir. Michael R.A. Abdelmalik (TU/e)
Title: T.b.a.
11 May 2026, 16:00 (T.b.A)
- Speaker: MSc. Tom Jacobs (CISPA)
Title: T.b.a.