Finman Tables Figures

Tables & Figures (finman-tables-figures)

When to trigger

• The main result is settled and must be made readable at a glance
• Tables are dense, over-decimaled, or bury the headline coefficient among a dozen columns
• An event-study / first-stage / portfolio-sort plot needs to carry the design visually
• You are preparing exhibits for the FMA→ScholarOne submission and want them house-style clean

The FM exhibit bar

FM's brand is "articles that people actually read," and that taste extends to the tables. The main causal or economic estimate should be findable in seconds, and — because FM weighs practical relevance — every headline exhibit should make the economic magnitude legible, not just the t-statistic. Finance house style permits significance stars but requires standard errors in parentheses; report the dependent-variable mean so a reader can scale the coefficient into a managerial quantity. The journal's "less weight on trivial robustness" stance also shapes the exhibit set: do not pad the main paper with specification-sweep tables — promote the one headline table and the one identification figure, and demote the rest.

Exhibit craft

1. One table for the headline. A referee should read the main estimate, its SE, N, and dependent-variable mean without flipping pages.
2. Standard errors always; stars optional. SEs in parentheses are load-bearing; state the clustering level in the note; if stars are used, define them.
3. Make the magnitude speak. Add a one-line economic-significance statement or column — a one-SD change in X moves Y by Z percentage points / dollars — so the practical relevance FM prizes is visible.
4. Figures carry the design. Event-study leads and first-stage relationships persuade better as clean vector figures with CIs than as prose.
5. Right precision and a lean set. Two to three significant figures; keep the main paper to the exhibits that earn their place and push sweeps to the internet appendix.

Execution bridge (StatsPAI / Stata MCP)

Generate exhibits from the fitted result, not by retyping numbers. Full map: execution-with-mcp. Financial Management is empirical corporate finance + asset pricing; corporate-causal chain (DiD/IV/RDD) plus the factor-zoo haircut for cross-sectional pricing.

• Tables: etable (multi-model) or did_summary_to_latex straight from the result_id.
• Figures: plot_from_result / enhanced_event_study_plot / event_study_table — axis units and the SE/clustering note baked in.
• Every note names the estimator + clustering and states the magnitude in interpretable units.

See a full fitted-result → exhibit chain in the JF execution walkthrough.

Checklist

• Main estimate readable in one table: coefficient, SE in parentheses, N, dep-var mean, controls flagged
• Standard errors reported everywhere; clustering level in notes; stars (if used) defined
• An economic-magnitude statement accompanies the headline result
• Identification figure present (event-study with CIs / first-stage / sorts) where relevant
• Notes make each exhibit self-contained (sample, units, clustering, controls, period)
• Precision sensible (2–3 sig figs); specification sweeps demoted to the internet appendix

Making the magnitude speak (FM's signature exhibit move)

Because FM weights practical relevance, the single highest-return exhibit edit is making the economic size legible. Three reliable ways:

1. A dedicated economic-significance column or panel — alongside the coefficient, report the effect of a one-SD change in the regressor as a percentage of the dependent-variable mean.
2. A managerial-unit translation in the note — convert the coefficient into basis points, dollars per firm, or percentage of payout, so a CFO or investor reads the size directly.
3. A back-of-envelope aggregate — where credible, scale the per-firm effect to a portfolio or market total, with the assumptions stated. A table that reports only t-statistics asks the referee to do this work; FM referees often will not, and "is this big enough to matter?" becomes the rejection.