I am an Assistant Professor of Econometrics at the University of Michigan.

My theoretical research interests are nonparametric approaches for statistical identification, estimation, and inference. My current focus is on developing geometric methods for causal inference.



ffg [at] umich [dot] edu

Research papers

Distributional synthetic controls (major revision requested, slides)

A path-sampling method to partially identify causal effects in instrumental variable models (revised, slides, this paper by Kilbertus, Kusner and Silva (2020) introduces a more efficient and robust (albeit slightly less general) practical implementation of the method from my paper)

Testability of instrument validity under continuous treatments (revision requested)

On the convergence rate of potentials of Brenier maps (2nd revision resubmitted)

Identifying multivariate models with binary instruments via cyclically monotone dynamics (revising)

Independent nonlinear component analysis (joint with Susanne Schennach, submitted)