The popular statistical computation and graphics system, R, offers a comprehensive suite of tools for data analysis. With its language, runtime environment, debugger, and access to system functions, R provides researchers and statisticians with a powerful platform to run and analyze programs stored in script files.
To get started with R, users can easily install the R package, which may require the installation of gcc-fortran for external package installation. Once installed, users can open a terminal and initiate an R session by simply typing a command.
During an R session, users have access to various features and functionalities. For instance, they can read the documentation about system file configuration, access online help, utilize the HTML browser interface for help, explore demos, and close the session when desired.
When closing an R session, a prompt will appear, allowing users to save their workspace, which includes user-defined objects and functions. The saved image is stored in a specific format and will be automatically reloaded when R is started again. Users can also manually save the workspace and load images during their session.
Upon starting R, its configuration is determined by several files. The site and user environment variable files are loaded first. The site file, controlled by the environment variable, defaults to a specific name. Similarly, the user file, specified by a separate variable, defaults to a specific name or location if not set. The Environment Variables R Documentation provides essential information about these variables.
Alternatively, users can set environmental variables from within their R session using the function. This allows for customization, such as setting the time zone variable to a specific value.
Furthermore, R can load a specific file containing R code for execution. Multiple files can be used, but only one is loaded in a specific order of preference. It is recommended to avoid loading packages or executing code that could hinder package upgrades and reproducibility.
For additional functionality, users can turn to the desolve repository, which offers a range of pre-built R packages. To request these packages, users can refer to the repository’s Git repository.
Installing packages within R is straightforward. Users can use the command to install packages directly from R. By default, packages are installed in a directory determined by a specific element returned by , which itself defaults to the value of an environment variable. However, users can change the installation directory using the ‘lib’ argument of .
To ensure the correct setup, it is recommended to use a local library and let pacman manage files residing under it. Users can confirm the existence and correctness of their user library within their R session using a specific command.
In addition to installing packages from within R, users can also install them from the command line. This provides flexibility, especially when rebuilding packages or selecting a specific mirror from which to download the packages.
For package updates, users can utilize , which is included with R, from the shell.
To optimize package installation, users can modify the Makevars file to set default make options. An optimized Makevars file example is provided for reference.
It is important to note that R does not offer a point-and-click graphical user interface for statistics or data manipulation. However, there are third-party user interfaces available, such as R Commander and Rattle, which provide a user-friendly experience for R users.
R Commander, available as an R package, is a popular user interface that requires the installation of tk. Users can start R Commander from within R using the library command.
Similarly, Rattle is a user interface focused on data mining. It can be installed easily from within R and relies on gtk2. Users can start Rattle from within R using the library command.
For advanced users, RKWard is an IDE developed by KDE, offering data import, browsing, statistical tests, and plots. Users can install rkward from the official Arch repositories.
Another widely used R IDE is RStudio, which is open-source and includes various conveniences like parentheses matching, tab-completion, and a spreadsheet-like data viewer. RStudio’s default layout features a four-pane view. However, users can customize the layout by creating a specific file with elevated privileges and modifying its contents accordingly.
In some cases, when starting RStudio, users may encounter an error related to openssl-1.1 package. To resolve this, they need to install the required package.
RStudio Server enables users to access a browser-based interface for running R on a remote Linux server. This brings convenience and flexibility for remote work scenarios.
KDE’s cantor notebook application also supports R, providing an integrated environment for scientific and statistical computing.
For those using the Visual Studio Code editor, R is supported through plugins, allowing for a seamless coding experience.
Another browser-based notebook, jupyter-notebook, also supports R when the IRkernel is installed.
While the numerical libraries that come with the R package do not have multithreading capabilities, users can replace the reference blas with an optimized BLAS library to significantly enhance computational speed for common R operations. OpenBLAS is a recommended option, and if using the regular r package from the extra repository, no further configuration is necessary.
For Intel processors, users can leverage the Intel Math Kernel Library (MKL), which not only provides multithreading capabilities but also includes specific optimizations for Intel processors. However, it’s important to note that these optimizations may affect standard R functionality for parallel processing. To utilize the MKL, users should first install intel-oneapi-mkl and then the r-mkl package.
To avoid repetitive prompts regarding the CRAN mirror for package installation or updates, users can set the mirror in the file. A default mirror, such as https://cloud.r-project.org/, is often recommended.
When exiting R, users are typically greeted with a prompt asking if they want to save the workspace image. While this may seem convenient, using workspace images can reduce code portability. To disable this prompt, users can add code that alters the default value of the argument to their file.
R continues to be a popular choice for statistical computation and graphics among researchers and statisticians. With its vast array of features, packages, and user interfaces, R offers a versatile environment for data analysis and research projects.
