Abstract:
The workshop will focus on recent advances in causal inference and causal mediation analysis. Causal inference is essential for comparative effectiveness research and causal discoveries from (large) observational data (including EHRs). Causal mediation helps us understand how an exposure or intervention works through different pathways. Both become considerably more complex in the presence of interference, on networks, with time-varying exposures, and in big data settings with many potential confounders and/or many potential mediators. Important issues to be addressed include the complication of causal inference in the presence of interference, causal inference on networks, causal mediation analysis in the presence of many mediators, variable selection, sensitivity analysis for uncheckable assumptions (including unmeasured confounders) with applications to ‘omics, mental health, education, networks, and more.
Program Information
Guest Speakers
Edoardo Airoldi, Millard E. Gladfelter Professor of Statistics & Data Science, Fox School of Business, Temple University, “Model-assisted design of experiments on social and information networks”
Jennifer Hill,Professor of Applied Statistics & Data Science, New York University, “Identifying Heterogeneous Treatment Effects with Bayesian Nonparametrics”
Luke Keele,Associate Professor of Statistics in Surgery, University of Pennsylvania, “Estimation Methods for Cluster Randomized Trials with Noncompliance: A Study of A Biometric Smartcard Payment System in India”
Fan Li, Associate Professor of Statistical Science, Duke University, “Introducing the overlap weights in causal inference”
Hongzhe Li, Professor of Biostatistics & Statistics, Perelman School of Medicine, University of Pennsylvania, “Inference for Individual Mediation Effects and Interventional Effects in Sparse High- Dimensional Causal Graphical Models”
Jasjeet Sekhon, Robson Professor of Political Science & Statistics, University of California at Berkeley, “Transfer Learning for Estimating Causal Effects using Neural Networks”
Michael Sobel, Professor of Statistics, Columbia University, “Between Causation and Association: Inference for the Role of Judge Attributes in EEOC Litigation Outcomes”
Elizabeth Stuart, Associate Dean for Education, Professor of Mental Health, Biostatistics, and Health Policy and Management, John Hopkins Bloomberg School of Public Health, “Confounders? Assumptions? Huh? Mediation analysis in applied research”
Eric Tchetgen Tchetgen, Luddy Family President’s Distinguished Professor & Professor of Statistics, The Wharton School, University of Pennsylvania, “Longitudinal Marginal Structural Models Estimation with Instrumental Variables: Identification and Multiple Robustness”
Stijn Vansteelandt, Professor of Statistics, Ghent University (Belgium) and the London School of Hygiene & Tropical Medicine, “Time-to-event analysis of randomized trials with repeatedly measured mediators: a re-analysis of the LEADER trial”
Organizers
Dr. George Michailidis and Dr. Michael Daniels