It was quite appalling to observe the most widely used and appreciated statistical analysis software, SAS, used to come at a premium price. Unless you had the luxury to study from an institute that had a SAS license you would have been devoid from the tool unless you join an organization which had the license. The matter was even worse for people willing to make a career shift by learning this powerful tool as even the individual license would cost too much. Well SAS University has heard all our pleas, and we have a free SAS called the SAS U or the SAS University Edition. We’ll understand why SAS has come up with this free solution and also do a quick review of the product.
Why is SAS offering a free solution?
For many years, unlike the open source data analysis languages like R, Python which were free to use by anyone, SAS had maintained a huge subscription fee even for the individual licenses and made tons of money by selling software to large organizations. However, the popularity of SAS is steadily declining in academics where R/Python has become one of the preferred programming languages. This declining market share in academics is a big issue for SAS as students who graduate with expertise in R/Python would be the analysts of tomorrow and having worked in R/Python would prefer it over SAS.
Who is it for? How does it work?
The university edition of SAS (SAS U) can be used by anyone for non-commercial purpose. Yes, you don’t need to be a university student to download and install it!
SAS U runs as a virtual machine, hence, you first need to download and install the VirtualBox software offered by Oracle free of cost. You may also use VMware Player or VMware Fusion for the same. After downloading the SAS U, you just need to open the VirtualBox and install SAS U as a virtual machine. To start SAS U, then you just need to open VirtualBox and key in http://localhost:10080 in any of the web browsers. SAS U works on any browser and can be installed on Windows, Mac or Linux servers. You can also use it as a cloud through Amazon Web Services. The minimum system requirement is a 64-bit machine with at least 1 GB of memory. You don’t need internet to run SAS U. Please follow detailed installation instructions here.
What all is included in SAS University Edition?
In SAS U you can access SAS via SAS Studio. SAS Studio is a new interface for accessing SAS via a web browser. For anyone accustomed with traditional SAS windows, picking up on different windows of SAS Studio should not be much of a trouble.
The SAS U is quite powerful and has most of the components needed to perform statistical analysis. The following components are available for download along with SAS U:
- Base SAS – A powerful programming language that’s easy to learn and easy to use.
- SAS/STAT – Comprehensive, reliable tools that include state-of-the-art statistical methods.
- SAS/IML – A robust, yet flexible matrix programming language.
- SAS/ACCESS – Access data from different PC file formats.
What’s missing in the SAS University Edition?
Though the existing components of SAS U are sufficient for most of your data analysis, advanced users may still feel handicapped by the absence of the below components:
- SAS/ETS (Econometrics & Time Series)
- SAS/OR (Operations Research)
- SAS/QC (Quality Control)
Amongst all of the above, at first look, the biggest disappointment is the missing SAS/GRAPH. How are we supposed to do any analysis without plotting the data? However, the SAS U has ODS graphics so even though you can’t use PROC GPLOT our job is done pretty well with PROC SGPLOT. Another disappointment is the missing SAS/ETS, so all your time series analysis, ARIMA etc. can’t be done here. However, a workaround for this is the SAS on Demand for Academics, which though needs an internet connection to run but has SAS/ETS in it and your login credentials for the SAS U account works for this too!
Other things we miss are the SAS Enterprise Guide, Enterprise Miner or Visual Analytics. So your advanced machine learning algorithms like decision trees, random forests, etc. need to wait until SAS provide these modules.
What’s more to watch out for?
One of the important caveat to watch out for is the restricted RAM usage by SAS U. SAS U is designed to use only 1GB of your system RAM or a max of 2GB with 2 processors. This could dampen the performance for advanced users and justifies the basic tenet of SAS U, it is for non-commercial purposes only! However, from a learning point of view, we were able to work quite comfortably with large datasets (size ranging from 100 – 500 MB).
Another important thing to look out for is the community discussions. SAS is actively trying to create a community of SAS U users though it’s at the nascent stage and lacks the depth and breadth of topic discussion as you would see for R or Python.
We have always been supportive of free license software’s at least for the learning purpose, and are all excited to see this initiative from SAS University. Though some components are missing and there are some performance caveats, the current version should get you running with most of the tasks you need to perform for data analysis.