skills.rest
zhouziyue233
zhouziyue233COMMUNITY
Explore all skills by this author →
GitHub

rdd-analysis

Estimate causal effects with sharp and fuzzy RDD analysis around a cutoff.

6|1|Updated Apr 3, 2026
Search Tags:#robustness#econometrics#rdd#regression-discontinuity#manipulation-test#bandwidth-selection#local-linear
One-click install
npx skills add https://github.com/zhouziyue233/great-econometrics --skill rdd-analysis-zhouziyue233
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rdd-analysis
Source: https://github.com/zhouziyue233/great-econometrics/tree/main/skills/rdd-analysis
Command: npx skills add https://github.com/zhouziyue233/great-econometrics --skill rdd-analysis-zhouziyue233

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Regression Discontinuity designs (RDD) enable causal inference at a known threshold, supporting sharp and fuzzy designs around a cutoff. This skill provides a complete workflow with guidance on identification assumptions, bandwidth selection, local polynomial estimation, validity checks, and reporting standards to ensure credible empirical results.

Core Features & Use Cases

  • Visualize the discontinuity with a binned scatter plot before any regression to assess data support.
  • Select optimal bandwidth (IK or CCT) and estimate using local linear (or local polynomial) methods.
  • Estimate main RDD effects for both sharp and fuzzy designs, plus validity tests (McCrary density test, covariate balance checks, placebo cutoffs, bandwidth sensitivity).
  • Produce reporting-ready results and diagnostics suitable for academic papers and policy evaluations.
  • Apply to scenarios like policy thresholds, program eligibility designs, and geographic RDD variations.

Quick Start

Run a complete RDD analysis on a dataset with a running variable and a cutoff to obtain the main estimate, robustness checks, and diagnostic plots.

Frequently Asked Questions about rdd-analysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I perform a regression discontinuity design analysis with bandwidth selection and validity tests?▼

Regression discontinuity design analysis is performed by visualizing binned scatter plots, selecting optimal IK or CCT bandwidths, estimating local linear effects, and running McCrary density tests to ensure credible causal inference at the cutoff.

What is the difference between sharp and fuzzy RDD estimation?▼

Sharp RDD assumes treatment is strictly determined by the cutoff, while fuzzy RDD handles partial compliance where the probability jumps discontinuously, both estimable using local polynomial methods with appropriate bandwidth selection.

How do I run a McCrary density test for manipulation in RDD?▼

The McCrary density test checks for manipulation around the cutoff by examining whether the density of the running variable changes discontinuously, complementing covariate balance checks and placebo cutoffs for robust validity testing.

Can I use local polynomial regression for policy threshold evaluations?▼

Local polynomial regression is applicable to policy threshold evaluations, program eligibility designs, and geographic RDD variations, providing reporting-ready estimates and diagnostics suitable for academic papers.

What is the best way to check robustness in regression discontinuity designs?▼

Robustness in regression discontinuity designs is checked through bandwidth sensitivity analysis, placebo cutoff tests, and covariate balance checks, ensuring main effect estimates remain stable across different specifications.

Related Skills

Explore More Data & Analytics Skills→
sheehesheehe

rdd-analysis

Estimate local treatment effects at known cutoffs using regression discontinuity designs.

Community
Advanced
JunMA98JunMA98
3

scientific-visualization

Generate publication-ready CS visualizations with accessible color palettes and export formats.

Community
Intermediate
jkitchinjkitchin
34

python-plotting

Select and apply Matplotlib, Seaborn, or Plotly for Python visualizations.

Community
Intermediate
tondevreltondevrel
19

dowhy

Estimate causal effects from observational data using the identify-estimate-refute framework.

Community
Advanced