Microsoft R Fundamentals
Wrangling large amounts of data, producing publication-ready graphics and visualizations, and generating predictive machine learning models are all possible with R.
Designed by statisticians for statistical computing, R has become the go-to programming language for data science.
Microsoft bought Revolution Analytics in 2015 and offers an enhanced distribution of open source R called Microsoft R Open. The software allows R to utilize multiple cores, resulting in speedups for certain analysis operations.
Getting Started
R is a code-based program that you use by typing what you want it to do rather than clicking buttons. Your commands and their results can be typed either in the console window or the script window (which you can tab along with the console). When you type a command, RStudio checks it for syntax errors. Any errors are highlighted with a red squiggly line. A message explaining the error is displayed in the Error List window, which you can access from the View menu or by clicking on it.
R also keeps track of what you’ve typed, and lets you access the previous command by hitting the up or down arrow keys. That’s great when you forget what you just typed or need to look at it again. But it is a bit clumsy, so it’s best not to use the up or down arrows too often. For example, if you type 10 = 20, R will try to interpret it as a command, typo and all, before whinging and spitting out an error.
Installation
R is an open source programming language with a comprehensive community providing documentation and support. Microsoft R Open is an enhanced version of R that provides multithreaded performance and specialized packages for big data analytics. It is compatible with all scripts, packages and applications that use R-4.0.2 and is maintained by the R core team.
R has a command-line interface and an interactive graphical user interface called RStudio Desktop. The graphical interface is more intuitive and easy to learn for novices, and is available in both Windows and macOS.
The RStudio Desktop GUI is not required to run R but can improve the experience by making it easier and faster to work with. It also provides an integrated development environment (IDE) for R and a notebook that allows you to create and interact with the results of your analysis. The RStudio Desktop IDE is free and can be downloaded from the RStudio website for your operating system.
Functions
R is an interpreted programming language. It comes with a command-line interface and multiple third-party graphical user interfaces. It can also be accessed through high-level scripting languages like Python,[71] Perl,[72] Ruby, and Java.[73] It is also part of several statistical frameworks, such as Jamovi and JASP.
Microsoft offers a variety of proprietary functions for data science and machine learning at scale. These include functions for transforming, training and analyzing datasets, scoring, text and image analysis, and feature extraction. Some of these functions can be parallelized, while others cannot.
XLSTAT-R allows you to apply R procedures directly within XLSTAT dialog boxes in Microsoft Excel. This provides the flexibility of R with the power of XLSTAT’s dialog box user interface. The resulting solution is very intuitive and easy to use. You can even edit your own combined R and XLSTAT procedures using XML code templates. The result is a powerful and flexible interface that allows you to import data from flat files, statistical software, or databases.
Compatibility
Microsoft Compatibility Telemetry is a feature that collects technical data from your PC to improve the experience. However, it can cause high cpu usage on some PCs. To fix this issue, you need to disable the telemetry process.
After Revolution Analytics was acquired by Microsoft, some were concerned that the enhanced R distribution formerly known as Revolution R Open would be discontinued. But in January 2016, Microsoft announced it would continue to offer both Microsoft R and Microsoft R Open.
The CRAN website provides files to build R from source on Debian, Redhat and SUSE systems. It guides users through the process for each system, with documentation and README files providing details.
The installer for each version of R installs a personal library (which is kept separate from the installed packages) and a temporary directory to store compiled binaries. This way, when a minor or even major version change is released one can keep the existing installations of packages and the same configuration of R.