[Statlist] ETH Young Data Science Researcher Seminar Zurich, Virtual Seminar by Raaz Dwivedi, UC Berkeley, 25 September 2020

[Maurer Letizia]

Dear all

We are glad to announce the following talk in the virtual ETH Young Data Science Researcher Seminar Zurich

"StaDISC: Stable discovery of interpretable subgroups via calibration"
by Raaz Dwivedi, UC Berkeley

Time: Friday, 25 September 2020, 17:00-​18:00
Place: Zoom at https://ethz.zoom.us/j/92367940258

Abstract: In this talk, I will present some recent work where we introduce a novel methodology for Stable Discovery of Interpretable Subgroups via Calibration (StaDISC), with large heterogeneous treatment effects, for randomized experiments. Building on Yu and Kumbier's PCS framework, StaDISC was developed during our re-analysis of the 1999-2000 VIGOR study, an 8076 patient randomized controlled trial, that compared the risk of adverse events from a then newly approved drug, Rofecoxib (Vioxx), to that from an older drug Naproxen. On average, and in comparison to Naproxen, Vioxx was found to reduce the risk of gastrointestinal events but increase the risk of thrombotic cardiovascular events. Applying StaDISC, we fit 18 popular conditional average treatment effect (CATE) estimators for both outcomes and use calibration to demonstrate their poor global performance. However, we find that CATE methods are locally well-calibrated and stable, thereby enabling the identification of clinically interpretable patient groups with larger than (estimated) average treatment effects. External validation of the found subgroups provides further evidence for the promises of the proposed methodology.

Based on joint work with Briton Park, Yan Shuo Tan, Mian Wei, Kevin Horgan, David Madigan, and Bin Yu
arxiv preprint https://arxiv.org/abs/2008.10109 (in submission to international statistical review)

Best wishes,

M. Löffler, A. Taeb, Y. Chen

Seminar website: https://math.ethz.ch/sfs/news-and-events/young-data-science.html

	[[alternative HTML version deleted]]