A subset of the Oregon Health Insurance Experiment (OHIE) contains 12,229 individuals who satisfied the inclusion criteria and who responded to the in-person survey by October 2010. It has bee
Other experiment results include downstream analysis, ablation studies, and investigation of hyperparameter settings. The results are for the updated scNODE paper.
We derive new variance formulas for inference on a general class of estimands of causal average treatment effects in a randomized control trial. We generalize the seminal work of Robins and show that
