Professor Eric Tchetgen Tchetgen
Department of Statistics and Data Science
University of Pennsylvania
Eric Tchetgen Tchetgen is University Professor, Professor of Biostatistics in Biostatistics and Epidemiology and Professor of Statistics and Data Science, at Wharton, University of Pennsylvania. Before coming to UPenn, he was Professor of Biostatistics and Epidemiologic Methods with joint appointments in the departments of Biostatistics and Epidemiology at the T.H. Chan School of Public Health (Harvard). He received his BS in Electrical Engineering from Yale in 1999 and his PhD in Biostatistics from Harvard in 2006.
His primary area of interest is in semi-parametric efficiency theory with application to causal inference, missing data problems, statistical genetics and mixed model theory. He works on the development of statistical and epidemiologic methods that make efficient use of the information in data collected by scientific investigators, while avoiding unnecessary assumptions about the underlying data generating mechanism. He has won many prestigious awards during his career including co-winner of the Rousseeuw Prize for Statistics (2022), the Myrto Lefkopoulou Distinguished Lectureship (2020), and co-winner of the Society of Epidemiologic Research and American Journal of Epidemiology Article of the Year.
General Lecture
Wednesday, November 8, 2023, 4:00PM to 5:00PM
Title:
Abstract:
In this talk, I discuss recent advances at the nexus of Causal Inference, Semiparametric Theory and Modern Machine Learning in the gilded age of data science, with emphasis given to sound and responsible decision-making in Public Health and Medicine. I argue with several recent developments, that important scientific progress can be made more likely by fruitful interaction between these three disciplines than without; however, such success will require constant vigilance to ensure that key foundational principles underpinning each discipline are preserved and respected. For illustration, I consider the potential utility and limitations of several recent advances emblematizing this (un)holy union, including conformal predictive causal inference, Double-debiased machine learning, Credence and pitfalls of the so-called Deconfounder.
Technical Lecture
Thursday, November 9, 2023, 4:00PM to 5:00PM
Title:
Introducing the Forster-Warmuth Nonparametric Counterfactual Regression
Abstract:
Series regression estimation is one of the most popular non-parametric regression techniques in practice. The most routinely used series estimator is based on ordinary least squares fitting, which is known to be minimax rate optimal in various settings, albeit under stringent restrictions on the basis functions. In this work, inspired by the recently developed Forster-Warmuth (FW) regression, we propose an alternative nonparametric series estimator that can attain minimax estimation rates under strictly weaker conditions imposed on the basis functions, than virtually all existing series estimators in the literature. Another contribution of this work generalizes the FW-regression to a so-called counterfactual regression problem, in which the response variable of interest may not be directly observed (hence, the name “counterfactual’‘) on all sampled units. Although counterfactual regression is not entirely a new area of inquiry, we propose the first-ever systematic study of this challenging problem from a unified pseudo-outcome perspective. In fact, we provide what appears to be the first generic and constructive approach for generating the pseudo-outcome (to substitute for the unobserved response) which leads to the estimation of the counterfactual regression curve of interest with small bias, namely bias of second order. Several applications are used to illustrate the resulting FW counterfactual regression including a large class of nonparametric regression problems in missing data and causal inference literature, for which we establish conditions for minimax rate optimality. This is joint work with Yachong Yang and Arun kuchibhotla.