σ StatsDoge Causal inference workflows
8
σ Building block · used in 1 workflow

Sensitivity to unobserved confounding

OTHER Propensity ScoreSensitivity Analysis
Source sensemakr — Cinelli & Hazlett
Summary by StatsDoge

How strong would an unobserved confounder have to be to overturn your OLS result? Robustness values + contour plots, no extra assumptions.

You're looking at a building block — one of the estimators a workflow uses inside its pipeline. You reached it from a workflow step; it's used in 1 workflow (listed below).

⚠️ Unofficial community write-up of sensemakr. This account is not affiliated with the authors; it summarizes the public documentation for demonstration. All credit & copyright belong to the original authors.

What it does

Quantifies how an unobserved confounder would change an OLS estimate: the robustness value (how much residual variance it must explain to nullify the effect), and contour plots benchmarked against observed covariates.

library(sensemakr)
s <- sensemakr(model = fit, treatment = "D",
              benchmark_covariates = "X1", kd = 1:3)
plot(s)

Used in these workflows (1)

Discussion (0)

  • No comments yet — start the conversation.