// Use when assembling the AEA Data and Code Availability deposit for an AER, AER:Insights, or AEJ acceptance, writing the README, or auditing a replication package before the AEA Data Editor review. Implements the current AEA policy, including the February 2026 Data and Code Availability Policy.
| name | aer-replication |
| description | Use when assembling the AEA Data and Code Availability deposit for an AER, AER:Insights, or AEJ acceptance, writing the README, or auditing a replication package before the AEA Data Editor review. Implements the current AEA policy, including the February 2026 Data and Code Availability Policy. |
Prior to acceptance, every empirical, simulation, or experimental paper submitted to an AEA journal must provide data, code, and a complete README sufficient for an independent researcher to reproduce every reported result. The AEA Data Editor's Office now performs a computational reproducibility check before final acceptance, and a failed check delays publication by weeks or months.
This skill produces an AEA-compliant deposit on the first try.
Three pillars:
Materials must be deposited in an openly accessible trusted repository. The strongly encouraged repository is the AEA Data and Code Repository at openICPSR, which gives the Data Editor automatic access to draft deposits.
These forms are provided by the editorial office at the appropriate stage. Do not invent substitute forms.
replication-package/
├── README.md (or README.pdf)
├── LICENSE (commonly MIT or CC-BY for code; data per source license)
├── data/
│ ├── raw/ (original files as obtained, never modified)
│ ├── intermediate/ (cleaned analytic datasets)
│ └── codebook/ (variable definitions, source mapping)
├── code/
│ ├── 00_setup.do (or .R, .py)
│ ├── 01_clean.do
│ ├── 02_analysis.do
│ ├── 03_tables.do
│ └── 04_figures.do
├── output/
│ ├── tables/
│ └── figures/
└── docs/
├── data_appendix.pdf
└── computing_environment.txt
The AEA Data Editor's office publishes a template. Required sections:
One paragraph: what the package does, what paper it accompanies, what software is required.
For every dataset used, state:
If data cannot be deposited (proprietary, restricted, IRB), the author must:
A table of every file in data/, with columns: Filename | Description | Source | Notes.
Operating system, software version, packages, estimated runtime, peak memory. Example:
Software: Stata 18.0 MP, R 4.4.1
Stata packages: reghdfe (6.12), ftools (2.49), did (1.3.1), rdrobust (9.2)
R packages: fixest (0.12.1), did (2.1.2), modelsummary (2.1.1)
Hardware: standard laptop (8 GB RAM); full run completes in ~25 minutes.
What each script does and the order to run them. State the master script (e.g., run_all.do).
A literal step-by-step:
1. Download this archive and unzip to a working directory.
2. Open Stata 18; install dependencies via `setup/install_packages.do`.
3. Edit line 12 of `00_setup.do` to point to the working directory.
4. Run `do run_all.do`. Total runtime: ~25 minutes.
5. All tables and figures are written to `output/`.
A second table mapping each exhibit in the published paper to the script that produces it:
| Table/Figure | Script | Line in Script | Output File |
|---|---|---|---|
| Table 1 | 03_tables.do | 12 | output/tables/tab1.tex |
| Table 2 | 03_tables.do | 67 | output/tables/tab2.tex |
| Figure 1 | 04_figures.do | 8 | output/figures/fig1.pdf |
Adopt from day one. Retrofitting is expensive.
data/raw/foo.csv, never /Users/yourname/...setup script with ssc install ... , version; in R, use renv or groundhog; in Python, requirements.txt with exact versionslogs/version 18.0 at the top of every do-fileset seed before any randomizationset more off and other display tweaks in shared scriptsreghdfe over xtreg, fe for performance and clarityset.seed() at the top of every analysis scriptrenv::snapshot() to lock the package environmentread.csv and writing with write.csv — explicit data.table::fread / fwrite with column typesrequirements.txt with == pinsnumpy.random.default_rng(seed=...) for new-style RNG.py scriptIf yes to all: Approved, paper proceeds to publication. If no: Revisions required, with a public report.
Common failure modes flagged in published Data Editor reports:
If data is restricted:
When working from the AER-skills repository or plugin bundle, load only the relevant resource:
examples/replication-package-skeleton/templates/stata/, templates/r/, or templates/python/examples/aer-exemplars.mdDEPOSIT TYPE: <public-data | restricted-data | hybrid>
README STATUS: <complete | incomplete>
REPRODUCIBILITY CHECK: <self-tested clean / not yet>
RESTRICTED DATA HANDLING: <n/a | provenance documented>
NEXT SKILL: aer-submission
Use when selecting, implementing, or stress-testing the causal identification strategy for an empirical economics manuscript — difference-in-differences (including staggered designs), instrumental variables (including weak-IV-robust inference), regression discontinuity, synthetic control, or shift-share / Bartik. Apply before writing the introduction or results.
Use when drafting or rewriting the introduction of an economics manuscript targeted at AER, AER:Insights, or an AEJ, or when compressing an abstract to the mandatory 100-word limit. Implements the Keith Head / Bellemare five-paragraph formula and AER-specific formatting conventions.
Use when responding to a Revise & Resubmit decision from AER, AER:Insights, or an AEJ, and a point-by-point response letter plus aligned manuscript revisions are needed. Handles triage, the concede / clarify / push-back decision, and the response-letter format that editors actually read.
Use when constructing or revising regression tables, descriptive statistics tables, or figures for an AER, AER:Insights, or AEJ manuscript. Implements AER booktabs house style, the standard regression-table layout, and the figure-notes convention.
Use when the main empirical results exist but the manuscript lacks the robustness, heterogeneity, mechanism, and placebo checks that AER referees will demand. Apply after aer-identification and before aer-introduction so that the value-added paragraph can reference these tests.
Use when running the final pre-submission audit for an AER, AER:Insights, or AEJ manuscript — length, format, cover letter, per-author disclosure statements, file packaging, and routing among the AEA journal family. Apply immediately before clicking submit.