Agent Skills
› brycewang-stanford/Awesome-Journal-Skills
› car-tables-figures
car-tables-figures
GitHub用于生成符合《当代会计研究》(CAR)风格指南的实证论文附录，包括描述性统计、相关性、回归结果表格及实验均值表。确保自包含格式，强制变量定义附录，并支持通过MCP工具从分析结果自动构建图表，避免手动输入错误。
Trigger Scenarios
需要按照CAR风格指南格式化论文附录或图表
缺少变量定义的附录或实验结果的均值表
需要从Stata/StatsPAI结果自动生成合规的LaTeX表格和图形
Install
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill car-tables-figures -g -y
SKILL.md
Frontmatter
{
"name": "car-tables-figures",
"description": "Use when building the exhibits of a Contemporary Accounting Research (CAR) manuscript — descriptive\/correlation\/results tables for archival work, experimental cell-means and ANOVA tables, model figures, and appendices — formatted to the CAR Style Guide (each exhibit on a new page, variable definitions, footnotes not endnotes). Builds exhibits; it does not run the analysis (car-data-analysis)."
}
Tables & Figures (car-tables-figures)
When to trigger
- Exhibits are cluttered, not self-contained, or off CAR house style
- A variable-definitions appendix is missing (an archival reviewer staple)
- Experimental results lack a clean cell-means / contrast table
- You are paginating exhibits for the 50-page limit and online appendix
CAR Style Guide exhibit rules
CAR's format rules (author guidelines + CAR Style Guide) govern exhibits directly:
- Each exhibit (appendix, table, figure) begins on a new page.
- In-text notes are footnotes, not endnotes; tables use table notes to define every variable, sample, and significance convention.
- Pages are numbered sequentially after the abstract page; the manuscript itself is blind (no author identifiers in exhibit titles, file names, or footnotes).
- The main text is held to 30 pages within a 50-page overall limit; push large or supplementary exhibits into the online-only supplementary appendix rather than padding the main body.
- LaTeX authors may use Wiley's New Journal Design (NJD) template for accepted papers; confirm exact table/figure formatting against the current CAR Style Guide.
Build exhibits the reviewers expect, by tradition
- Archival / capital-markets. Provide (1) a variable-definitions appendix with exact construction and data sources — this is mandatory in spirit under the Data Integrity policy; (2) a descriptive-statistics table (N, mean, median, SD, key percentiles), noting winsorizing/truncating; (3) a correlation table (Pearson/Spearman); (4) results tables with coefficients, standard errors (clustered), and the SE-clustering and fixed-effects structure stated in the notes. Report economic magnitude, not just stars.
- Experimental. Provide a cell-means table (means, SDs, n per condition), the ANOVA/contrast table, and manipulation-check results; a figure of the predicted interaction aids interpretation.
- Analytical. Present propositions/results cleanly; figures should illustrate comparative statics or equilibrium regions, with parameter assumptions noted.
Self-containment
Every exhibit must be readable without the text: title, sample/period, variable definitions, units, and the meaning of each significance marker all appear in the notes.
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. CAR is archival/empirical accounting; the DiD / IV / RDD chain serves its causal designs around reporting and regulation.
- 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
- Each exhibit starts on a new page; footnotes (not endnotes) used throughout
- Variable-definitions appendix present (archival) with sources and construction
- Descriptive, correlation, and results tables report clustered SEs and FE structure
- Experimental cell-means, ANOVA/contrasts, and manipulation checks tabulated
- Economic magnitudes (not only significance) reported
- Exhibits blind; supplementary material routed to the online-only appendix
- Numbers reconcile exactly with the code repository and the text
Anti-patterns
- Stars without magnitudes; endnotes where CAR requires footnotes.
- Undefined variables or a missing definitions appendix.
- Author identifiers leaking into exhibit titles or file names.
- Padding the main body past 30/50 pages instead of using the online appendix.
Exhibit pass for Contemporary Accounting Research
Use this as a second-pass capability check. First lock the accounting construct, setting, identification or theory, and disclosure/market/organizational implication; then test whether the manuscript addresses accounting reviewers who expect accounting-specific constructs, credible design, and contribution to reporting, auditing, tax, or governance debates.
- 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 The Accounting Review for US flagship breadth, JAR for Chicago-style accounting research, JAE for economics/accounting interface; 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】tables/figures/appendices; each on a new page?
【Variable definitions】appendix present, sources/construction stated?
【Results table】clustered SEs, FE structure, magnitudes shown?
【Experiment】cell means / ANOVA / manipulation checks tabulated?
【Style】footnotes (not endnotes), blind, page budget (30/50) ...
【Next step】car-writing-style
Version History
- 1839142 Current 2026-07-05 12:48
