rddlisted

# Regression discontinuity An opinionated RD workflow grounded in a read canon (references/canon.md, current as of 2026-07-28): the Cattaneo-Titiunik Annual Review of Economics survey and its applied companion, the Cattaneo-Keele-Titiunik guide, which includes a real failed design this skill uses as its refusal template. Deliverable: the recommendation with its citation, the R estimation and diagnostics code, and a methods paragraph. Refresh path: run litreview on the method since the canon date, then propose additions to references/canon.md as flagged addenda. ## Design shapes and the case that anchors each Each canonical case is a precedent a methods section can cite; details.md says what each teaches. | Design shape | Canonical case | Marketing analogue | What kills it | |---|---|---|---| | Age or tenure eligibility rule | Card, Dobkin, Maestas 2008 | trial expiry, anniversary status rolloff | something else switches at the same threshold | | Agency-measured score, sharp | Hansen 2015 | churn-score retention offers | reps or managers override the rule off-cutoff | | Heaped or rounded score | Almond et al. 2010; Barreca et al. 2011, 2016 | spend thresholds recorded in whole dollars | excess mass at round values that the density test misses | | Share crossing a fixed bar | Lee, Moretti, Butler 2004 | seller badge above an on-time-delivery rate | predetermined covariates already jump in the smallest window | | Admission cutoff, fuzzy first stage | Hoekstra 2009 | lead-sc