From restud-skills
Guides selection, implementation, and stress-testing of causal identification strategies for REStud manuscripts: DID, IV, RDD, synthetic control, shift-share, RCT.
How this skill is triggered — by the user, by Claude, or both
Slash command
/restud-skills:restud-identificationThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- The empirical core is OLS + controls and a referee will flag selection-on-observables
REStud was founded in 1933 to advance both theoretical and applied economics, and that balance is its signature: unlike a purely empirical top-5 outlet, REStud weighs design-based causal work and structural/theory-consistent identification on equal terms. So the bar is a state-of-the-art causal design, OR — for a structural/theoretical-empirical paper — a rigorous identification of the model's parameters with the identifying assumptions stated explicitly and mapped to data features. The empirical contribution can itself be a new identification strategy; in that case the design is the paper and must be defended to the hilt. Editorial assignment reflects the breadth: handling Joint Managing Editors span IO/applied econometrics (Jan De Loecker), micro theory / mechanism design (Antonio Penta), and behavioral/information economics (Jakub Steiner) — write the identification so the right one of them sees a rigorous argument. Identification is not a section to be footnoted; if the design is fragile, no prose, controls, or sample size rescues it.
Is treatment plausibly random by design (RCT, lottery)?
├── Yes → analyze as an experiment; pre-register / PAP where applicable
└── No → identification comes from variation
├── Sharp threshold in a running variable → RDD (sharp / fuzzy)
├── Policy hits some units over time, not others → DID
│ ├── One treatment date → canonical 2x2 DID
│ └── Staggered adoption → Callaway-Sant'Anna / Borusyak-Jaravel-Spiess / Sun-Abraham
├── Endogenous regressor + plausibly exogenous shifter → IV
│ ├── Shifter x exposure shares → shift-share / Bartik
│ └── Single instrument, F < 50 → weak-IV-robust inference (AR)
├── One / few treated aggregate units → synthetic control
└── Structural parameter of interest → restud-theory-model (state identifying assumptions)
csdid / did), Borusyak-Jaravel-Spiess imputation, de Chaisemartin-D'Haultfœuille, or Sun-Abraham.rdrobust).rdplot with the binning stated.restud-theory-model; supply counterfactuals.【STRATEGY】DiD | IV | RDD | SCM | shift-share | RCT | structural
【MODERN ESTIMATOR】yes / no / which
【DIAGNOSTICS REPORTED】[...]
【INFERENCE】cluster-robust / AR / wild bootstrap / permutation
【RED FLAGS】[... or "none"]
【NEXT SKILL】restud-robustness (or restud-theory-model if structural)
npx claudepluginhub brycewang-stanford/awesome-journal-skills --plugin restud-skillsStress-tests causal identification and measurement strategy for REStat manuscripts (DID, RD, IV, shift-share, measurement-error problems).
Stress-tests quasi-experimental identification strategies (DID, IV, RDD, DML) for economic research manuscripts against modern estimators and reporting standards.
Stress-tests causal identification arguments (RCT, DiD, RD, IV, shift-share) for AEJ: Applied manuscripts, ensuring designs meet the journal's credibility bar before exhibits are finalized.