Scope

The Journal of Statistical Software (JSS) publishes open-source software and corresponding reproducible scholary articles discussing all aspects of the design, implementation, documentation, application, evaluation, comparison, maintenance and distribution of software dedicated to improvement of state-of-the-art in statistical computing in all areas of empirical research. Open-source code and articles are jointly reviewed and published in this journal and should be accessible to a broad community of practitioners, teachers, and researchers in the field of statistics.

Types of Papers

JSS aims at publishing articles describing comprehensive open-source implementations of broad classes of statistical models and procedures or computational infrastructure upon which such implementations can be built. JSS evaluates and publishes software and articles as a unity. Full reproducibility is mandatory for publication and the source code is published along with the article. Articles describing a specific software implementation must present the scope of the software and the underlying models or concepts such that statistically literate readers can understand what the software does. A careful comparison with other open-source implementations of similar models or procedures should highlight the capabilities of all implementations and the corresponding advantages or disadvantages. A description of the design principles and the actual implementation is at the heart of an JSS article. The general analysis workflow must be illustrated by an enlighting non-trivial case study.

JSS publishes special issues on topics in statistical computing and statistical software. Software described by articles compiled in a special issue should address similar models and procedures or be dedicated to a specific area of application. Potential guest editors of such a special issue are welcome to send proposals to the editors-in-chief.

Descriptions of open-source software targeting less broad problems, for example compact yet sophisticated code to estimate novel models using existing software infrastructure, can be published as code snippets in JSS. Articles of this type are typically rather short.

Book reviews published in JSS are overseen and solicited by the book editor. Similar to book reviews, JSS also considers reviews and structured comparisons of closed-source statistical software, ie software distributed under a licence incompatible with the GNU General Public Licence. To ensure fairness and unbiasedness, potential authors of reviews and software comparisons should contact the editors-in-chief prior to engaging in such work. Assessments or comparisons of open-source software are not within the scope of this section. Authors interested in a comparison of open-source software packages are encouraged to team up for comprehensive and fair comparison and review and submit their joint manuscript as a regular JSS paper. One example of such a collaborative effort includes the paper State of the Art in Parallel Computing with R.

JSS does NOT publish methodological innovations in computational statistics. Novel models or estimation procedures should be published elsewhere prior to the submission of an article describing an implemention thereof to JSS. A subsequent publication of a corresponding software package in JSS is possible but requires a substantial software innovation. In contrast, JSS does not publish supplementary materials to methodological papers, such as implementations of rather specific models or algorithms (or accompanying documentation, algorithmic details, or simulations). Software WITHOUT a corresponding article documenting and illustrating the implementation and application of this software can not be published in JSS.

Code

Code can be in any interpreted or compiled high-level language, such as S/R, SPSS/Matrix, SAS/IML, MATLAB, Python, and Mathematica, Fortran, C, C++, Java, Common Lisp etc. Submissions of software written in languages requiring non-standard interpreters or compilers must include all tools necessary to run the code in the submission. To enhance accessibility authors of MATLAB software should generally test the use with Octave. If possible, the Octave compatibility should be pointed out in the manuscript - if not, the requirement for MATLAB should be justified in the manuscript. Although JSS publishes the source-code alongside with an article, it does NOT serve as a software repository for software distribution. It is not possible to update code or articles once they are published. If at all possible, software should also be available in a structured format from a standard open-source repository.