“Dual IV: A Single Stage Instrumental Variable Regression”

Dr Krikamol Muandet - Research group leader at the Max Planck Institute for Intelligent Systems, Tuebingen, Germany
Data Science

Date: November 19th 2019
Location: Eurecom - Eurecom

Abstract: In this talk, I will present a novel single-stage procedure for instrumental variable (IV) regression called DualIV which simplifies traditional two-stage regression via a dual formulation. We show that the common two-stage procedure can alternatively be solved via generalized least squares. Our formulation circumvents the first-stage regression which can be a bottleneck in modern two-stage procedures for IV regression. We also show that our framework is closely related to the generalized method of moments (GMM) with specific assumptions. This highlights the fundamental connection between GMM and two-stage procedures in IV literature. Using the proposed framework, we develop a simple kernel-based algorithm with consistency guarantees. Lastly, we give empirical results illustrating the advantages of our method over the existing two-stage algorithms. Joint work with Arash Mehrjou, Si Kai Lee, and Anant Raj Preprint: https://arxiv.org/pdf/1910.12358.pdf Bio: Krikamol Muandet is a research group leader at the Max Planck Institute for Intelligent Systems, Tuebingen, Germany. His research interests lie at the intersection of kernel methods, Hilbert space embedding of distributions, counterfactual inference, causal inference, and learning theory (website: http://krikamol.org).

Permalink: https://www.eurecom/seminar/92315