Revision Rebuttal (joe-rebuttal)
When to trigger
- A JoE decision letter arrived with referee reports and a revision request
- Referees challenge an assumption, a proof step, or the adequacy of the Monte Carlo
- You need a response-letter strategy that satisfies technical specialists
- You are about to resubmit and need to know what the current portal expects
The JoE revision reality
JoE referees are specialists who read methods papers closely, and review is single-anonymized (they know who you are). Reports are usually detailed and technical: a missing regularity condition, a gap in an asymptotic argument, a Monte Carlo that does not stress the assumptions, or an under-credited competing method. The editors make the final call after weighing the (minimum two) referees. Treat the revision as making the theory and simulations airtight, not as a rhetorical exercise. Note: resubmissions over one year old can re-trigger the USD $75 fee — budget for it.
Response-letter strategy
-
Restate each point in your own words before answering, so the referee sees you understood it.
-
Answer technically, with evidence. For a contested assumption: prove it is primitive, weaken it, or show the result is robust to its failure via Monte Carlo. For a proof gap: supply the missing lemma. For "the simulation is too narrow": add DGPs at the boundary and report size/power.
-
Point to exactly where in the manuscript each change lives (section, theorem, table number).
-
Concede gracefully where the referee is right; push back precisely (with a citation, a counterexample, or a derivation) where they are not. Never hand-wave.
-
Separate referees: a per-referee section, plus a short cover note to the editor summarizing the main changes.
Common JoE referee objections → responses
| Objection |
Response |
| "Assumption A is high-level / not verifiable" |
Replace with a primitive condition, or verify A for a leading model class |
| "The asymptotic argument has a gap" |
Supply the lemma; cite the empirical-process / ULLN tool used |
| "Variance estimator not shown consistent" |
Prove consistency or supply a valid bootstrap with theory |
| "Monte Carlo too favorable" |
Add heavy-tailed / dependent / boundary DGPs; report size + size-adjusted power |
| "How does this differ from [method]?" |
Add an explicit analytic + Monte Carlo comparison; sharpen positioning |
| "Illustration over-claims" |
Reframe as a demonstration of the method, not an applied finding |
Resubmission logistics
- Resubmit through the live official route. If the Editorial Express resubmission flow applies, use database
je; if the route points to Editorial Manager, follow that workflow.
- Check whether the fee re-triggers and stage proof of payment if the live route requests it.
- Provide a clean revised manuscript plus the response letter; keep the Elsevier formatting and ≤250-word abstract intact.
- Update the reproducible artifact (
run_all, estimator code, seeds) to match any new Monte Carlo.
Anti-patterns
- Drafting the response letter before the revised manuscript exists
- Rhetorical answers ("we respectfully disagree") with no derivation, citation, or new simulation
- Conceding a point in the letter but not changing the paper (or vice versa)
- Ignoring one referee because another was more positive — editors weigh all of them
Output format
【Decision】R&R / major / minor
【Per-referee plan】R1: [points→responses]; R2: [...]
【Theory changes】assumptions weakened / lemma added / variance proof
【Monte Carlo changes】new DGPs / boundary cases / size-power added
【Manuscript pointers】where each change lives
【Resubmission】live route checked; fee re-trigger checked? [Y/N]
【Next step】joe-submission preflight on the revised PDF
