From tar-skills
Runs and reports estimation for TAR manuscripts: matches estimator to design, handles fixed effects and standard-error clustering, runs robustness checks, and assembles data-authenticity/code-access documentation.
How this skill is triggered — by the user, by Claude, or both
Slash command
/tar-skills:tar-data-analysisThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- The sample is built and it is time to estimate and report
tar-methods: OLS with high-dimensional fixed
effects (firm, year, industry-year) for panel associations; DiD / staggered-DiD with a modern
estimator for adoption shocks; 2SLS for endogenous regressors; RDD for threshold settings; logit/
probit/Poisson/Tobit for limited or count outcomes (e.g., restatement, going-concern, fraud).TAR requires authors to enable confirmation of data authenticity, with differentiated rules:
Code/data sufficiency is part of the submission and acceptance requirements (待核实 whether a named public repository deposit is mandated at acceptance). Build clean, commented scripts from raw extract to every table now — not after the R&R.
【Estimator】OLS-HDFE / staggered-DiD / 2SLS / RDD / logit-Poisson ...
【Fixed effects & clustering】... ; SE level ...
【Focal measure】construction + alternative proxy: pass/issues
【Robustness】alt measures / samples / placebo / pre-trends ...
【Economic magnitude】coefficient means ... in accounting terms
【Data authenticity】public-db / abstracted / private — code & description ready? yes/no
【Open issues for reviewers】...
【Next step】tar-contribution-framing
npx claudepluginhub brycewang-stanford/awesome-journal-skills --plugin tar-skillsRuns and reports empirical archival analysis for JAR manuscripts: standard-error clustering, endogeneity, construct measurement, robustness battery, and reproducible data-and-code package.
Guides running and reporting empirical analysis for CAR manuscripts: selects the right estimator by design type (archival, experimental, analytical), enforces robustness checks, and prepares artifacts required by CAR's Data Integrity & Code Sharing Policy.
Runs and reports empirical analysis for JAE manuscripts: builds archival samples, specifies fixed effects and clustered standard errors, executes identification (DiD, IV, matching), and demonstrates robustness.