Biostatistics Dept. Seminar | Space-Time-Meta Causal Inference: A Weighting Perspective
Biostatistics Department Seminar
Title: Space-Time-Meta Causal Inference: A Weighting Perspective
Abstract: Causal inference with spatial, temporal, and meta-analytic data commonly defaults to regression modeling. While widely accepted, such regression approaches can suffer from model misspecification, pose interpretability challenges, and obscure how the available data are aggregated to produce effect estimates. In this presentation, we discuss a unified perspective grounded in weighting and optimization that subsumes and extends common regression procedures. For spatial data, we show that specifying the error structure in linear models is equivalent to imposing a particular covariate structure with appropriate regularization. For temporal data, we clarify how two-way fixed effects and related estimators aggregate information across time to construct causal contrasts. For meta-analytic data, we develop a diagnostic tool for detecting measurements whose covariate distributions meaningfully deviate from the target population, suggesting when it might be prudent to exclude them from the analysis. Viewing these disparate regression methods through a unified weighting lens simplifies estimation, clarifies how evidence is combined, and yields transparent diagnostics for causal inference in complex data settings across medical care and beyond.
Speaker
José Zubizarreta is Professor in the Department of Health Care Policy at Harvard Medical School, Professor in the Department of Biostatistics at the Harvard T.H. Chan School of Public Health, and Faculty Affiliate in the Department of Statistics at the Faculty of Arts and Sciences at Harvard University. His research focuses on the development of statistical methods for causal inference to advance health care and public policy research.
Zoom Registration
If you would like to join via Zoom, please register here.
2025-2026 Monday Seminar Series
All seminars are held at 12:05 PM via Zoom and onsite. View all seminar information here.