From eer-skills
Stress-tests causal identification designs (DiD, IV, RDD, experiment) for EER manuscripts, ensuring credibility before finalizing exhibits.
How this skill is triggered — by the user, by Claude, or both
Slash command
/eer-skills:eer-identificationThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- A causal claim rests on OLS + controls, or TWFE on staggered timing
EER publishes broadly across empirical economics, so identification is judged on credibility legible to a general reader: the mapping from variation in the data to the causal object must be explicit, the key assumption stated, and the most obvious threat pre-empted. Because review is single-anonymized, the referee is often a methods expert in your exact design — modern, design-appropriate estimators and honest inference are expected. Report standard errors and confidence intervals (EER house style; do not lean on significance stars — see eer-tables-figures). Match the size of the causal claim to what the design supports.
Clustering at the level of treatment assignment; with few clusters use wild-cluster bootstrap. Pair this skill with
eer-robustnessfor the specification/sample battery.
A migration paper uses a staggered visa-liberalization rollout. A weak version runs TWFE and reports a wage effect with stars. An EER version re-estimates with Callaway–Sant'Anna, shows flat leads and a dynamic post path, reports the effect as -1.4% local wages (s.e. 0.5, illustrative), runs Rambachan–Roth sensitivity, and states the estimand is the effect on incumbents in receiving regions — not a national average. The general-interest lesson (how labor supply shocks transmit to local wages) is named so a non-migration economist sees the point.
【Branch】DiD / IV / RDD / experiment
【Variation→object mapping】one sentence
【Key assumption】stated + the main threat pre-empted
【Design evidence】[pre-trends / first-stage F / density test / balance]
【Inference】SEs/CIs; clustering level; few-cluster fix?
【What it does NOT identify】[...]
【Next step】eer-theory-model (if a mechanism is needed) or eer-robustness
npx claudepluginhub brycewang-stanford/awesome-journal-skills --plugin eer-skillsStress-tests empirical identification strategies (DID, IV, RDD, event study, structural estimation) for economics manuscripts. Evaluates credibility and economic interpretation before drafting results.
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.
Helps refine identification arguments for IER manuscripts: structural parameter ID, causal design, or theory tightness. Does not run estimation.