liveBook · Manning

Rdrobust

As explained in Causal Inference for Data Science

rdrobust is a software package for regression discontinuity estimation using robust local polynomial methods.

rdrobust estimates a regression discontinuity effect at a threshold using local linear or local polynomial regression. It favors explicit, interpretable formulas over opaque machine-learning models. section 11.2.5

A practical workflow uses the package alongside exploratory and sensitivity checks:

  • Plot the discontinuity with rdplot before estimating the effect. section 11.2.5
  • Run rdrobust(Y, time_points) or its R equivalent to obtain an estimated effect and confidence interval. section 11.2.5
  • Treat the time window and interval carefully: a reported estimate of 770.05 had a confidence interval from −11.42 to 1551.52. section 11.2.5

A time-based discontinuity

For an outcome Y measured around an event time T, rdrobust estimates the jump at T. Comparing alternative time windows shows why a seemingly better interval can reflect choosing a window with knowledge of the synthetic answer. section 11.2.5

Listing 11.10 (R) Nonlinear estimation with the rdrobust function.
summary(rdrobust(Y, time_points))

Inspect the R call and returned summary for nonlinear estimation with rdrobust. section 11.2.5

How Causal Inference for Data Science covers it

How deep it goes
Explains plotting, invocation, confidence intervals, and local-linear reasoning without fully specifying every implementation detail.
Examples
Uses a synthetic time threshold, rdplot, varying time windows, and R and Python rdrobust calls.
What it assumes
Assumes familiarity with regression discontinuity and regression while motivating local-polynomial estimation from linear regression.
What you can do after
Apply it via rdrobust to estimate a discontinuity effect and inspect its confidence interval.
How it connects
Connects regression discontinuity, local linear regression, local polynomial regression, kernels, and confidence intervals.
Cover of Causal Inference for Data Science
liveBookmanning.com
Read Causal Inference for Data Science on liveBook Get the book at manning.com

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Causal Inference for Data Science. The section links point to the parts of the book it draws on.