Content and procedure

R is the de facto standard language for statistical computing and has thus also gained widespread acceptance in the field of data science, in which statistical analyses plays a key role.The relatively smooth learning curve is a big advantage, but the true strenght of R lies in the variety of extension packages available, e.g. on the official Comprehensive R Archive Network (CRAN). The variety of available methods of statistical learning is not comparable to any other software.

Bundling your R scripts into an R package has many advantages: Not only does it increase the chance of you reusing your own code, but it also encourages you to write a tidy and well documented code, which makes it much easier to comprehend even months and years after it was written. It thus also promotes the exchange of code between people and system platforms.

In this workshop we will develop an R package from scratch. We will examine the structure of R packages before writing our own code and build a package from it. Along the way we will take care to document not only the functions, but also the package itself adequately.

AT_Arne Kaldhusdal

Short Facts

  • Trainer: Arne Kaldhusdal
  • Language: English
  • 16th of April 2018
  • 10:00 – 17:15
  • Data Hub, Sapporobogen 6-8, 80637 München

Educational goals

Understanding why to develop R packages

Creation of a corresponding function documentation

Compiling R packages


Arne Kaldhusdal

is a statistician who has worked intensively with R for almost ten years, both in the academic context and in a variety of data science projects in various industries. He works as a data science expert at Alexander Thamm GmbH.


  • Participants should have practical experience in working with R.
  • Each participant should bring his own computer.
  • Each computer should have access to the Internet.
  • R should be pre-installed on every computer.
  • The R packages devtools, testthat and roxygen2 should be pre-installed on each computer.
  • Ideally, LaTeX should also be installed (e.g. MiKTeX on Windows systems).
  • Rtools should also be installed.

Are you interested in this workshop?