Data and Code Policy

JEPR asks referees to assess the evidence behind a claim, not only its description. This policy applies to every submission that reports numbers, tables, figures, experiments or simulations, and to results obtained by computation.

At submission

Authors provide the data and the analysis code that produce the reported results, either as submission files or through a private link. Where the data are simulated or synthetic, the scripts that generate them are provided instead, and the text states clearly that the evidence is simulated or synthetic. Referees receive these materials together with the manuscript.

Reproduction instructions and verification

Include a README identifying the main results to reproduce, the required files, software dependencies and versions, execution commands, random seeds where relevant and approximate execution costs. Identify reused data, exclusions and protocol deviations. Referees record which results they independently checked, which they could not check and why. Restricted-data arrangements must provide a justified alternative route for assessment.

For selective prediction and abstention, distinguish empirical risk estimates from proved guarantees. A guarantee requires a suitable theorem, explicit assumptions and an implementation satisfying those assumptions. Claims of equivalence require a justified margin and suitable analysis; lack of statistical significance alone is insufficient.

On publication

The data and code are deposited in a public repository that assigns a persistent identifier (for example Zenodo, OSF, Figshare or Software Heritage) and are cited in a Data and code availability statement in the article. We recommend CC BY 4.0 for data and an OSI-approved licence (such as MIT or Apache 2.0) for code.

Restricted data

Where data cannot be shared because of privacy, consent or third-party licences, authors explain the restriction at submission, describe how qualified researchers can request access, and provide the code and, where possible, aggregated or synthetic data that reproduce the analysis. The editor decides whether the arrangement is sufficient for review.

Formal results

Proofs are given in full in the article or its appendices. Computer-checked proofs, model checkers and enumeration scripts used to establish a result are deposited in the same way as code.

Example statement

Data and code availability: the data and code supporting this article are available in Zenodo at https://doi.org/10.5281/zenodo.XXXXXXX under the CC BY 4.0 and MIT licences.