Princeton University

School of Engineering & Applied Science

Permutation tests in the presence of confounders

Rina Foygel Barber, University of Chicago
Sherrerd Hall 101
Monday, December 10, 2018 - 12:30pm




When testing for dependence between two random variables X and Y, using a permutation test is no longer straightforward if we would like to control for the effect of confounding variables, i.e. testing conditional independence rather than marginal independence. If the relationship between one variable X and the confounder Z is known, then it's possible to resample X and see if apparent correlation between X and Y persists (Candes et al's conditional randomization test). We propose a permutation version of this strategy, permuting the values of X in a way that respects its dependence on the confounder Z, and assess the robustness of this approach to misspecification in the model linking X and Z.

This work is joint with Thomas Berrett, Richard Samworth, and Yi Wang.



Rina Foygel Barber is an Associate Professor in the Department of Statistics at the University of Chicago. Before starting at U of C, she was a NSF postdoctoral fellow during 2012-13 in the Department of Statistics at Stanford University, supervised by Emmanuel Cand├Ęs. She received a PhD in Statistics at the University of Chicago in 2012, advised by Mathias Drton and Nati Srebro, and a MS in Mathematics at the University of Chicago in 2009. Her research focuses on developing and analyzing estimation, inference, and optimization tools for structured high-dimensional data problems, particularly false discovery control in sparse regression problems. She also collaborates on modeling and optimization problems in image reconstruction for medical imaging.


This seminar is co-sponsored by S.S. Wilks Memorial Lecture series and the Korhammer Lecture Series