SAS represents one of the oldest data science tools. Developed in the 1960s, SAS is designed with a unique “library” structure that differentiates it from other data tools. This structure makes it particularly easy to work with extremely large datasets even when you have limited RAM or ability to subset data.
The primary advantage of SAS are as follows:
- SAS can work with virtually any quantity of data.
- Large and active online community.
- Easy integration with SQL syntax.
The primary disadvantages of SAS are:
- Complicated native syntax. PROC and DATA steps are very different from more modern data science tools, and it takes time to get used to these.
- Occasionally unhelpful error codes.
- Difficulty working with Excel data.
I typically only use SAS when I work with somewhat large data, usually greater than 20 gigabytes. For small jobs, the syntax just takes too long to write when the data is small and other tools are available.