Simon Urbanek

Michael J. Kane, Simon Urbanek

Reproducible document standards, like R Markdown, facilitate the programmatic creation of documents whose content is itself programmatically generated. While these documents are generally not complete in the sense that they will not include prose content, generated by an author to provide context, a narrative, etc., programmatic generation can provide substantial efficiencies for structuring and constructing documents. This paper explores the programmatic generation of reproducible by distinguishing components than can be created by computational means from those requiring human-generated prose, providing guidelines for the generation of these documents, and identifying a use case in clinical trial reporting. These concepts and use case are illustrated through the listdown package for the R programming environment, which is is currently available on the Comprehensive R Archive Network (CRAN).

Michael J. Kane, Bryan Lewis, Sekhar Tatikonda et al.

Linear regression models depend directly on the design matrix and its properties. Techniques that efficiently estimate model coefficients by partitioning rows of the design matrix are increasingly popular for large-scale problems because they fit well with modern parallel computing architectures. We propose a simple measure of {\em concordance} between a design matrix and a subset of its rows that estimates how well a subset captures the variance-covariance structure of a larger data set. We illustrate the use of this measure in a heuristic method for selecting row partition sizes that balance statistical and computational efficiency goals in real-world problems.