NeurIPS 2020

Sense and Sensitivity Analysis: Simple Post-Hoc Analysis of Bias Due to Unobserved Confounding


Meta Review

The paper makes an important contribution to the important problem of sensitivity analysis. The paper is very well written and presented, and the technical contribution is interesting and would surely prove useful for practitioners of modern causal inference methods. The reviewers made many useful suggestion which I trust the authors will incorporate into their work. I also suggest citing [1], which focuses on the issue of calibrating the sensitivity analysis parameters to the observed confounders. [1] Hsu, Jesse Y., and Dylan S. Small. "Calibrating sensitivity analyses to observed covariates in observational studies." Biometrics 69.4 (2013): 803-811.