Colloquium

Department of Statistics

Statistics Colloquium
January 31 (Fri) 11 a.m.
Alavi Commons Room, 6625 Everett Tower

Literate statistical software development with Rstudio

David Zeitler
Associate Professor of Statistics
Grand Valley State University

The Rstudio integrated development environment (IDE) for R has come a long way in a very short time. It provides not only a great environment to edit and run your R code, but also includes tools to bring good software engineering to your R code development. These tools include literate programming using Sweave and/or Rmarkdown and modern source control using GIT.

Sweave is a tool that allows the analyst to embed R code in Latex documents. When R is run, tables and graphs are automatically inserted into a Latex document. The Latex document is then automatically updated if data change, which allows for truly up-to-date reports.

While Sweave creates Latex documents, R Markdown creates web reports and HTML documents. This enables web pages to be automatically updated whenever underlying code or data changes. R Markdown is also readily useable even on windows machines where Sweave and LaTeX are generally problematic.

GIT is the latest open source tool for managing source code. It tracks changes over time and manages efforts of multiple developers or even your own multiple development tracks.

We will use a classroom example of repeated measures analysis in R and my Empirical Spectral Test research work to explore how these tools can be used in class work, research and consulting work.

All statistics students are expected to attend.

Past colloquiums