Agent Skills
› brycewang-stanford/Awesome-Journal-Skills
› jae-tables-figures
jae-tables-figures
GitHub为JAE期刊生成符合Elsevier规范的表格与图表，包括变量定义、样本构建、描述性统计、回归结果及事件研究图。确保数据自包含、可复现，并规范标准误聚类、固定效应及显著性标记，提升学术呈现质量。
Trigger Scenarios
需要生成JAE格式的回归表或描述性统计表
缺失变量定义或样本构建说明
需绘制事件研究或双重差分动态图
原始回归输出未格式化且缺乏统计细节
Install
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill jae-tables-figures -g -y
SKILL.md
Frontmatter
{
"name": "jae-tables-figures",
"description": "Use when building the exhibits for a Journal of Accounting and Economics (JAE) manuscript — variable-definition appendix, sample-construction table, descriptive statistics and correlations, regression tables with clustered standard errors, and event-study\/DiD figures — in Elsevier house style so each exhibit is self-contained and reproducible."
}
Tables & Figures for JAE (jae-tables-figures)
When to trigger
- Regression output is pasted in raw and not yet reader-ready
- Tables do not state the sample, the SE clustering, or variable definitions
- A reviewer cannot tell which coefficients are the test of the hypothesis
- You need a variable-definition appendix or a sample-waterfall table
The standard JAE exhibit set
Archival JAE papers carry a recognizable table sequence; build it deliberately:
- Variable definitions (often an appendix): every variable, its construction, and the data source (Compustat/CRSP/I/B/E/S/Execucomp/DealScan/Audit Analytics). Reviewers check this first.
- Sample construction: the waterfall from population to final N with counts at each exclusion.
- Descriptive statistics: N, mean, median, SD, and key percentiles; note winsorization (e.g., 1%/99%).
- Correlation matrix: Pearson (and often Spearman) correlations among main variables.
- Main regressions: the hypothesis tests, with the coefficient of interest visually identifiable.
- Cross-sectional / mechanism tables: partitions showing the effect concentrates where the friction is severe.
- Robustness tables: alternative proxies, specifications, placebo/falsification, identification diagnostics.
Reporting conventions reviewers expect
- Report coefficients with t- or z-statistics (or standard errors) and state the SE clustering (firm, or two-way firm-and-year) in the table note.
- Indicate the fixed effects included (firm, year, industry) in each column.
- Use consistent significance markers and define them in the note; report economic magnitude, not only stars.
- Every table note must make the exhibit self-contained: sample, period, units, FE, clustering, and what the key coefficient tests.
Figures
- Event studies: plot cumulative abnormal returns around the information event with confidence bands.
- DiD dynamics: plot event-time coefficients to show parallel pre-trends and the post-treatment effect.
- Comparative statics (analytical papers): plot the model's predicted relations. Keep figures clean, labeled, and grayscale-legible; Elsevier renders single-column native format.
Elsevier house style
JAE follows Elsevier author-date (Harvard) referencing and numbered sections (1, 1.1, 1.1.1). Number tables/figures in citation order, reference each in the text, and keep the in-text discussion interpreting—not merely repeating—the numbers. Remember the manuscript also requires Highlights (2-5 bullets, ≤125 characters each) summarizing the findings.
Execution bridge (StatsPAI / Stata MCP)
Generate exhibits from the fitted result, not by retyping numbers (the usual source of
body-vs-appendix drift). Full map: execution-with-mcp. JAE is empirical accounting with an economics lens; treat identification and weak-IV-robust inference as the binding constraints.
- Tables:
etable(multi-model columns) ordid_summary_to_latexstraight from theresult_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 effect size in interpretable units.
See a full fitted-result → exhibit chain in the JF execution walkthrough.
Checklist
- Variable-definition appendix with data sources
- Sample-construction waterfall table
- Descriptives + correlation matrix; winsorization noted
- Main tables: key coefficient identifiable; t-stats/SEs shown
- FE and SE clustering stated in every table note
- Mechanism and robustness tables present
- Event-study/DiD figures with pre-trends and bands
- Every exhibit self-contained; Highlights drafted
Anti-patterns
- Raw software dumps with unlabeled columns.
- No clustering/FE disclosure in table notes.
- Stars without magnitudes.
- Undefined variables or missing data-source attribution.
- DiD with no event-time figure to support parallel trends.
Exhibit pass for Journal of Accounting and Economics
Use this as a second-pass capability check. First lock the economic mechanism, accounting setting, identification or model, and market/contracting consequence; then test whether the manuscript addresses accounting-economics reviewers who expect economics discipline, identification, and market or contracting implications.
- 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 editrows so the next pass can patch the manuscript directly. - Neighbor test: compare against JAR for accounting research breadth, TAR for accounting flagship breadth, Review of Accounting Studies for archival/accounting focus; 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.mdfor volatile rules and name the one unresolved fact that could change the recommendation.
Output format
【Exhibit list】vars / sample / descriptives / corr / main / mechanism / robustness / figures
【Each main table note】sample, period, FE, SE clustering, sig. definitions
【Figures】event-study CARs / DiD dynamics / comparative statics
【Highlights drafted?】2-5 bullets ≤125 chars
【Next step】jae-writing-style
Version History
- 1839142 Current 2026-07-05 13:25
