TAMIDS Seminar Series: Abhishek Chakrabortty: Causal Inference for Treatment Effects from Observational Data: An Overview

Speaker: Abhishek Chakrabortty, Ph.D., Assistant Professor, Department of Statistics and College of Arts and Sciences, Texas A&M

Abstract: Causal inference is a fundamental and one of the most active areas of research, with a vast literature spanning across decades in both statistics and machine learning, and with wide applications across disciplines, including biology, computer science, econometrics etc. The causality literature is fairly diverse as well, with varied goals and associated methods. In this tutorial, I will focus on the causal inference literature for treatment effect estimation, specifically the average treatment effect (ATE), from observational data – which are increasingly popular in the modern “big data” era due to their rich and large-scale nature. I will first introduce the basic principles and philosophy guiding such causal problems, including counterfactual reasoning and the “potential outcome” framework. I will then discuss the key challenges with observational data (as opposed to randomized trials which are the gold standard) due to issues of confounding and selection bias, and the key causal assumptions one needs to make for valid inference. Thereafter, I will discuss specific strategies for identification and estimation of ATE from such data. In particular, I will discuss three popular methods: inverse probability weighting (IPW), regression adjustment, and augmented IPW (so-called doubly robust or double machine learning) estimators. Their advantages/disadvantages will be discussed in detail and some of their theoretical and empirical properties will be discussed briefly as well.

Prerequisites and logistics: This tutorial is generally intended for an audience with a background in statistics, machine learning or related areas. Familiarity with basic concepts and tools of probability/expectations and statistical inference/regression methods is expected. No prior background in causal inference is required. This tutorial will be in the form of a two-hour lecture session, with room for questions in between. There is no hands-on session in this tutorial, but examples will be used to facilitate understanding throughout the lecture. Bringing your own device is not necessary.

Biography:  Dr. Chakrabortty is an Assistant Professor in the Department of Statistics at Texas A&M University (TAMU). Dr. Chakrabortty received his Ph.D. in Biostatistics from Harvard University, and bachelors & Masters from the Indian Statistical Institute. Prior to joining TAMU, he was a postdoctoral fellow in the Department of Statistics at the University of Pennsylvania. His methodological research broadly focuses on robust and efficient inference with high dimensional and/or incomplete observational data. Some of his main research interests include semi-supervised learning, high dimensional inference, missing data and causal inference etc., with applications in the analysis of large-scale data from modern biomedical studies. His work is currently funded by the National Science Foundation, and he is also a recipient of the Texas A&M Institute of Data Science (TAMIDS) Career Initiation Fellow Award.

This tutorial will be held on March 29, 2023 from 02:00pm-04:00pm in Blocker Building-Room 220. Attendance is limited to 40 participants. To register for the seminar, please click here.

If you have any questions, please contact Diego Rodriguez, TAMIDS Assistant Director for Educational Programs.