4 ms·
Its funny to me because for a lot of R users programming has very little to do with their use of R i.e. hypothesis testing using libraries like lme4 for mixed m
by SubiculumCode 8y ago
Its funny to me because for a lot of R users programming has very little to do with their use of R i.e. hypothesis testing using libraries like lme4 for mixed models.
- sampo 8y agoIf the language doesn't matter, why do you use R and not GLIMMIX in SAS?
- SubiculumCode 8y agoI didn't mean it didn't matter, just that it is not at the forefront. The real reason is the open source licensing, the huge number of quality or advanced packages for stat analysis and graphing.
- 4thaccount 8y agoFiguring out how to use R is straight forward. It is much like any interpreted scripting language and there is a huge amount of blog posts, stack overflow questions, books, YouTube videos, and courses on how to do things. It is free. SAS is expensive, the training is a joke, figuring out how to use it is bizzare. I could go on and on. I did a training at work on how to use R to a group using SAS to get them off of it and save us a significant amount of money. SAS seemed to be installed on the server and getting the client side to work on my computer took lots of effort between me and IT. SAS does have easy access to the databases we use, but it was really easy to get R database access with ODBC. In my opinion the gulf between Matlab and Python is much smaller than SAS and R. Granted, I'm not a statistician or expert in either R or SAS. I'm sure SAS does have some use cases where it shines (maybe where you need a really clear audit trail), but overall it seemed to be more confusing than it is worth.