Please Note: This seminar will be given online.
University of Pennsylvania
Link to join seminar: Hosted on Zoom
Removing Strong Data Assumptions in Causal Inference via Large-Scale Mathematical Programming
Many traditional and newly developed causal inference methods rely on strong assumptions about the data distribution and/or data collection procedure, and if those assumptions do not hold, they may either be inapplicable or suffer from low power.
In this seminar talk, we use two examples to illustrate how large-scale mathematical programming can aid in conducting valid and powerful causal inferences without imposing strong data assumptions, and how these methodological advances benefit health data analytics and infectious disease research.
If time permits, we also briefly discuss some ongoing projects and future directions in mathematical-programming-based methods in causal inference.