mgsci-tables-figures - SKILL.md Agent Skill

name: mgsci-tables-figures description: Use when building exhibits for a Management Science (INFORMS) manuscript — propositions and numerical-illustration figures for analytical papers, or result/identification tables and effect plots for empirical papers — keeping notation clean, exhibits self-contained, and the page budget tight for the invited-revision limit. It builds exhibits; it does not run the analysis (mgsci-data-analysis).

Tables & Figures (mgsci-tables-figures)

When to trigger

Exhibits are cluttered, notation-heavy, or not self-explanatory
A reviewer cannot read a result off a table/figure without hunting the text
You are over the page budget and exhibits are bloated
You need to present proofs/comparative statics or estimates in house style

Management Science prefers short, focused papers and warns that excessive length and notation density slow review. Every exhibit should earn its space and stand on its own.

Analytical lane

Propositions/theorems are numbered and stated cleanly in the text; defer long proofs to the appendix (which does not count toward the invited-revision page limit).
Comparative-statics figures are where the managerial insight becomes visible — plot the optimal policy / equilibrium quantity against the key primitive, with axes, parameter values, and the takeaway labeled.
Notation tables (a symbol glossary) help when the model is symbol-heavy; keep notation minimal and consistent throughout.
Numerical-example tables should report the parameter ranges and confirm the qualitative result holds across them.

Empirical lane

Descriptives / correlation table with clear variable definitions.
Main results table: coefficients with standard errors, the clustering noted, sample size, and fit; mark significance consistently.
Identification exhibits: event-study plots, parallel-trends checks, first-stage strength, RDD continuity plots — the design's credibility shown, not just asserted.
Effect plots: interaction/marginal-effect plots so the magnitude is readable; report practical magnitude, not only stars.

House style and self-containment

Use author-year style consistently in notes and captions.
Each table/figure has a number, a descriptive title, and a note defining every symbol, abbreviation, and the SE/clustering — readable without the body text.
Keep formatting to the journal's exhibit conventions; verify against current submission guidelines.

Page budget discipline

Invited revisions must fit within 47 pages double-spaced (25 lines/page) or 32 pages at 1.5 spacing (33 lines/page), 11-pt font, 1-inch margins — the online appendix is excluded. Move long proofs, secondary robustness tables, and extended derivations to the online appendix so the main exhibits stay essential.

Execution bridge (StatsPAI / Stata MCP)

Generate exhibits from the fitted result, not by retyping numbers (the usual source of main-text-vs-online-appendix drift). Full map: shared-resources/empirical-methods/execution-with-mcp.md.

Tables: etable (multi-model) or did_summary_to_latex straight from the result_id; figures: plot_from_result / enhanced_event_study_plot with axis units and the SE/clustering note baked in.
Keep the decisive exhibits in the page-limited body; route the rest to the online appendix. See a full fitted-result → exhibit chain in the JF execution walkthrough.

Anti-patterns

A wall of nested notation no glossary explains.
A results table the reader cannot interpret without the text.
Interaction effects reported only as coefficients, never plotted.
Padding the counted main body with exhibits that belong in the online appendix.

Exhibit pass for Management Science

Run this as a concrete capability pass. First lock the decision problem, formal or empirical engine, managerial lever, and generality claim; then test whether the manuscript addresses OR/MS reviewers who expect a generalizable decision model, credible empirical leverage, or algorithmic insight with managerial consequence.

Primary move: For every table or figure, state the object, sample/case base, uncertainty display, and one sentence the exhibit proves for this venue.
Decision ledger: return claim / evidence / blocker / next edit rows so the next pass can patch the manuscript directly.
Sibling comparison: compare against Operations Research for method-first optimization, Marketing Science for marketing models, Organization Science for organization-theory mechanisms; if the neighboring outlet has the stronger audience claim, recommend re-routing before polishing.
Verification floor: before submission-ready advice, re-open resources/official-source-map.md for volatile rules and name the one unresolved fact that could change the recommendation.

Worked micro-example (illustrative) and exhibit pushback

A platform-competition paper's main empirical table reports a commission elasticity of −0.4 (illustrative) but only as a starred coefficient. Because Management Science readers span many departments, a finance or operations co-reviewer cannot judge the magnitude. The fix: add a marginal-effect plot showing commissions falling from 15% to 9% as search costs halve, with axes and the managerial takeaway labeled, and move the long proof of the comparative-static result to the online appendix (excluded from the invited-revision cap). Now both the analytical insight and the empirical magnitude read off the exhibits without the body text.

"I cannot interpret this table without the text." Add a self-contained note defining every symbol, the SE/clustering, and the sample; an exhibit must stand alone for a cross-department reader.
"The interaction is only a coefficient." Plot the marginal effect so the practical magnitude is visible, not just the star.
"The main body is over the page cap." Offload secondary robustness tables and long proofs to the online appendix; confirm the current cap against the author guidelines.

Output format

【Lane】analytical / empirical
【Core exhibits】[list, each self-contained]
【Notation/definitions】glossary + table notes complete: yes/no
【Identification/comparative-statics shown】yes/no
【Page budget】main body within invited-revision limit; appendix offloaded: yes/no
【Next step】mgsci-writing-style