From ier-skills
Organizes robustness checks for IER papers by threat to load-bearing assumption, without running regressions. Helps structure responses to referee concerns.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ier-skills:ier-robustnessThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- The headline number is suspected to hinge on a functional-form choice (CES/log/quadratic) or a single calibrated parameter
IER referees reward robustness that is tied to the load-bearing assumption identified in ier-theory-model/ier-identification. A list of twenty re-runs reads as defensive; three checks that each retire a specific threat to the result read as confident. Build the robustness section as a threat map:
| Threat to the result | The check that retires it | What "passing" looks like |
|---|---|---|
| Functional form drives it | Re-do under an alternative class (e.g., non-CES preferences, semiparametric) | Sign/magnitude survives; if it moves, you bound the movement |
| One calibrated/external parameter drives it | Vary it over a defensible range; report the headline as a function of it | The qualitative result holds across the range, or you state the threshold |
| Sample/period drives it | Drop influential subsamples, split the period, exclude outlier units | Estimate stable; no single unit/period is pivotal |
| Specification search | Show the result across a transparent grid of reasonable specifications | The result is not a lucky cell (a specification curve, not cherry-picking) |
| Inference is too generous | Cluster-robust / wild-cluster bootstrap / HAC as the design demands | CIs widen honestly and the conclusion still holds |
| Numerical fragility (structural) | Re-solve on a finer grid, alternative solver, tighter tolerance | Counterfactual stable to solution method |
At a theory-leaning journal, "robustness" is broader than re-running regressions. For a theory paper it means showing the result survives perturbing the load-bearing assumption (the perturbation test from ier-theory-model) — that is the robustness section. For a structural paper it means three distinct things: (a) economic robustness — does the counterfactual survive an alternative preference/technology specification; (b) parameter robustness — does it survive varying the externally-set parameters over a defensible range; and (c) numerical robustness — does it survive re-solving on a finer grid with a different solver. An IER referee distinguishes these and will not accept (b) as a substitute for (c) or (a). Map each explicitly.
A quantitative spatial model reports a welfare gain of 4.5% from removing a trade barrier. The threat map flags the externally-set migration elasticity as the prime suspect. Instead of one re-run, the authors report the welfare gain as a function of the elasticity across its literature-supported range — say 3.6% to 5.4% — and note that the qualitative conclusion (positive, economically meaningful) holds throughout. They then re-solve the model on a doubled grid to confirm the number is not a discretization artifact. This reads as confident and bounded; a single "robust to alternative elasticity" footnote would have invited exactly the objection it tried to dodge.
With a ≤50-page ceiling, the robustness section competes for space with the contribution itself, so be ruthless about placement. The main text holds the one or two checks that retire the threats tied to the load-bearing assumption — these are part of the argument, not an addendum. Everything else (alternative control sets, secondary subsamples, placebo variants) goes to the online appendix, referenced in one sentence. A main text crowded with low-value re-runs signals defensiveness; a main text with two decisive checks and a clear pointer to the appendix signals an author who knows which threats matter.
The defining feature of robustness at a rigor journal is that honest fragility, bounded, beats false robustness. If a check moves the headline, the worst response is to omit it; the second worst is to bury it; the best is to report it and bound it ("the effect falls by a third under the alternative specification but remains positive and significant at conventional levels"). IER referees have seen every way of hiding a fragile result, and discovering a concealed one — especially via the replication deposit — is close to fatal. Treat the robustness section as the place where you demonstrate you have already tried hardest to break your own result and it survived.
【Journal】International Economic Review
【Skill】ier-robustness
【Result under test】the headline number/sign
【Threat map】ranked threats → the decisive check for each
【Main-text checks】the 1–3 tied to the load-bearing assumption
【Parameter range】headline reported as a function of the uncertain parameter? [Y/N]
【Honest fragility】any check that moves it, with the movement bounded
【Inference】clustering/bootstrap/HAC matched to design? [Y/N]
【Verdict】robust / bounded-and-stated / fragile
【Next skill】ier-tables-figures
npx claudepluginhub brycewang-stanford/awesome-journal-skills --plugin ier-skillsOrganizes robustness checks for JEEA manuscripts around threats a general-interest referee would raise, ensuring headline results are stable to specification, sample, inference, and assumption perturbations.
Builds robustness suites for AEJ: Applied manuscripts to show headline estimates survive specification, sample, and inference choices.
Builds a robustness suite for REStat manuscripts: tests whether headline estimates survive specification, sample, measurement, identification, and inference alternatives.