"Causal Inference with Continuous Exposures"
Much of the literature on estimating causal effects concerns discrete exposures. However, continuous exposures such as air pollution, antibody response, or hours of job training abound in many scientific areas. In this talk, I will give a tutorial on causal inference with continuous exposures, using a dataset on county-level air pollution exposure as a working example. In particular, I will provide an overview of my recent research on nonparametric inference for causal dose-response curves. I will also discuss recent research by others on inference for personalized and incremental interventions.