From mgsci-skills
Executes and reports analysis for Management Science manuscripts: proves analytical results or estimates/validates empirical models, with replication package preparation.
How this skill is triggered — by the user, by Claude, or both
Slash command
/mgsci-skills:mgsci-data-analysisThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
- Results are ready to be derived/estimated and reported
Because Management Science is bimethodological, "analysis" means proving/computing results in the analytical lane or estimating/validating in the empirical lane. Both are held to their Department's rigor bar.
| Data structure / claim | Estimator |
|---|---|
| Causal effect, observational | DiD / IV / RDD / event study; cluster-robust SE |
| Causal effect, controlled | Experiment: randomization + manipulation/attention checks |
| Panel with unit heterogeneity | Fixed/random effects; clustered SE |
| Mechanism / structural primitives | Structural estimation tied to the model |
| Count / limited dependent variable | Poisson/NB, logit/probit, Tobit as fits |
Management Science enforces a Code and Data Disclosure Policy (effective June 1, 2019; revised April 20, 2026). Authors must create an AsCollected project page and provide its URL during submission. Accepted papers with numerical or computational work must provide data, programs, and details sufficient to permit replication before production, or an approved alternative disclosure plan for proprietary/sensitive data. Provide a master script, version-pinned environment, README, data sources, and a clear disclosure plan from the start.
The journal runs analytical and empirical work through different reviewer pools, and the evidence bar differs by lane:
| Element | Analytical lane | Empirical lane |
|---|---|---|
| Core claim | Theorem/proposition with a complete, checkable proof | Causal estimate with credible identification |
| Inference | Comparative statics with signed intuition | Standard errors clustered at the design level |
| Robustness | Relax key assumptions; show what breaks | Alternative specs, subsamples, placebo where available |
| Reproducibility | Self-contained proofs in the appendix | Master script + README that regenerates every reported number |
A frequent desk-to-reviewer failure is mixing lanes: an empirical paper that leans on an un-validated structural assumption, or an analytical paper whose "numerical illustration" silently does the real work. Name the lane, then clear that lane's bar before any prose or submission recommendation.
Run this as a concrete capability pass. First lock the decision problem, formal or empirical engine, managerial lever, and generality claim; then test whether the manuscript addresses OR/MS reviewers who expect a generalizable decision model, credible empirical leverage, or algorithmic insight with managerial consequence.
claim / evidence / blocker / next edit rows so the next pass can patch the manuscript directly.resources/official-source-map.md for volatile rules and name the one unresolved fact that could change the recommendation.【Lane】analytical / empirical
【Core result】proofs verified OR estimator + identification
【Comparative statics / inference】sign, intuition, clustered SE ...
【Robustness】relaxed assumptions / alt specs / subsamples ...
【Disclosure package】master script + README + sources: regenerates all main results? yes/no
【Next step】mgsci-contribution-framing
npx claudepluginhub brycewang-stanford/awesome-journal-skills --plugin mgsci-skillsExecutes and reports analysis for M&SOM manuscripts: proves structural results, runs numerical studies, estimates effects, and ensures replicability per INFORMS policy.
Guides selection and defense of analytical or empirical methods for Management Science manuscripts, matching the question to the appropriate Department standard.
Executes and reports analysis for Production and Operations Management manuscripts: proofs, numerical studies, empirical identification, and operational validation.