Michael Sobel is a professor of Statistics at Columbia University. His research interests lie primarily in the area of causal inference, where he has published papers on the subjects of mediation, interference, longitudinal causal inference using fixed effects models, meta-analysis, compliance, and causal inference for fMRI experiments, in which massive amounts of time series data are collected for subjects under varying experimental conditions. In addition to extending his work on fMRI, he is also working on interference in observational studies using fixed effects models, and he is working to develop some new estimands for counterfactual inference more broadly.