[This article was first published on Bart Rogiers - Sreigor, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.
In short: R is a free intuitive programming language that is used by practitioners in a plethora of academic disciplines. Therefore, it is on the cutting edge, and expanding rapidly. It creates stunning visuals, works seamlessly together with LaTeX, has really good online documentation and the community is unparalleled. [1, 2, 3, 4] A week or two ago, I sent an email to some of my colleagues to convince them to have a look at the capabilities of R. I decided to put my little R advertisement online, and here it is (already updated, since I’m learning every day more things that R can do). Let’s try to convince you now to join the club..
The R homepage, and the Comprehensive R Archive Network (CRAN) are the first places that you want to start browsing after deciding to learn R. A basic R introduction can also be found there, and the software sources, binaries and packages can be retrieved from these pages as well.
GUI’s for standard statistical analysis
Command-line programming is a big threshold to overcome by most people. A graphical user interface (GUI) should lower the threshold enough to be overcome. Probably the easiest GUI offering access to different statistical techniques through a drop-down menu and spreadsheet functionality for the data tables, is Rcommander. All basic statistical methods are in there, and an introduction is available on the web. So if you just need to perform some statistics on your data, this is the way to go.
Command-line programming is a big threshold to overcome by most people. A graphical user interface (GUI) should lower the threshold enough to be overcome. Probably the easiest GUI offering access to different statistical techniques through a drop-down menu and spreadsheet functionality for the data tables, is Rcommander. All basic statistical methods are in there, and an introduction is available on the web. So if you just need to perform some statistics on your data, this is the way to go.
Other, similar GUI’s exist, like Deducer and Rattle, but more exotic things are around too, that for instance apply the concept of visual programming, like Red-R and RAnalyticFlow.
More useful once you start writing your own scripts are simple code editors with syntax highlighting, that are interfaced with or built around R. An example is Tinn-R, which I have been using in Windows until now. I will be moving to RStudio however, since it seems very promising, and is cross-platform software.
Graphics capabilities
An overview of the graphics capabilities of R can be found in the R Graph Gallery. I also recently watched some interesting ideas of Hadley Wickham on the future of R graphics here. GUI’s for creating graphics exist as well, see for instance GrapheR.
An overview of the graphics capabilities of R can be found in the R Graph Gallery. I also recently watched some interesting ideas of Hadley Wickham on the future of R graphics here. GUI’s for creating graphics exist as well, see for instance GrapheR.
Packages
You can get a good idea on the packages that are available through the CRAN task views. Besides the standard programming and statistical functionality in R, I have already used several geostatistical packages, k-means and model-based clustering, artificial neural networks, and linear mixed modelling. I also know of several advanced geostatistical and inverse modelling tools that are being developed for it (e.g. multiple point geostatistics and an implementation of DREAM). There are even a few hydrological packages. All these come with extensive help files, so it is not too hard to figure out things that you have never used before.
You can get a good idea on the packages that are available through the CRAN task views. Besides the standard programming and statistical functionality in R, I have already used several geostatistical packages, k-means and model-based clustering, artificial neural networks, and linear mixed modelling. I also know of several advanced geostatistical and inverse modelling tools that are being developed for it (e.g. multiple point geostatistics and an implementation of DREAM). There are even a few hydrological packages. All these come with extensive help files, so it is not too hard to figure out things that you have never used before.
R can also be integrated with several GIS applications, as for instance Quantum GIS with manageR. The more advanced SAGA GIS functions can be called from R with RSAGA for performing terrain analysis for instance.
High performance computing, in the Cloud!
Reproducible research with R, LaTeX and Sweave
Several packages are available for parallel computing with R. The CRAN Task View gives an overview of the activity and development concerning HPC. Making use of multiple cores on your laptop is pretty easy with these packages. Moreover, a number of cloud computing services that support R are becoming available. An example is cloudnumbers.com.
Sweave integrates R with LaTeX, the most advanced scientific typesetting system. R code can be incorporated in your reports and papers, and is automatically executed to produce all your tables and graphs when you export your file to a pdf.
R is definitely a strong competitor for SAS, SPSS, Statistica, … but certainly also for other high-level programming languages like Matlab or Python! It is completely free, open source & cross-platform software, and has a vastly growing user base. SciLab and Octave are comparable languages, but should only be preferred for maximum Matlab compatibility in my opinion. Some comparisons can be found here and here. Sage is an integration of many open source packages, including R.
To leave a comment for the author, please follow the link and comment on their blog: Bart Rogiers - Sreigor.
R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.