SAS v R

noetsi

Fortran must die
#21
Bryan from the original article "SAS has become the undisputed market leader in commercial analytics space."

"SAS is commercial software. It's not cheap and still beyond reach for the majority of the professionals (in individual ability). Nevertheless, it holds the largest market share in Private Organizations."

what you said.


"R is not SAS. It doesn't try to be SAS. That article is stupid from the beginning when it says "R is the Open source counterpart of SAS." It isn't. That statement doesn't even make sense. They're not even remotely doing the same thing (SAS has a database engine, security settings, GUI interface, extensions, and so on; it's an application that can be used for data processing and statistics). R is a programming language that can be used to do a myriad of things, obviously designed for stats."

I agree. SAS is a commercial product employed largely by corporations [SAS's bread and butter is not statistics its forms of operations research these days even though its original creators were focused on statistics]. R I believe is primarily used by academics and high end users who do cutting edge analysis [in my humble experience few private companies do such because they don't believe what they get out of it is worth the cost. That might or might not be true, but corporate decision makers believe it is which is all that matters. Again that is only my opinion].

I am definitely an analyst not an academic or engineer [which seems humorous to say given that I earned 4 graduate degrees in social sciences, but clearly true].
 
Last edited:

bryangoodrich

Probably A Mammal
#22
"SAS has become the undisputed market leader in commercial analytics space."
Where's the evidence? Because Gartner says so? What's their methodology? How do they define "commercial analytics space" how did they compare competitors? Empirical claims require empirical support.
 

noetsi

Fortran must die
#23
I don't think there is any systematic data. I think it is based on their experience and those they talk to much as my comments are. There probably are surveys that ask companies what statistical/analytical software they use.
 

Jake

Cookie Scientist
#24
There is data.
There is data.
There is data.
There is data.
There is data.
There is data.
http://r4stats.com/articles/popularity/
You cannot keep ignoring this and spewing this unfounded **** about SAS being more popular than R in general.
Are there specific pockets of industry where SAS is more popular than R? Almost certainly. It looks like you work in such a pocket. But to say that SAS is more popular than R in general flies in the face of the evidence.
 

Dason

Ambassador to the humans
#25
Right now if I search for jobs on the ASA job board there are an equal number of SAS and R postings. If I search for jobs on stackoverflow within 100 miles of me there are 36 R jobs and 0 SAS jobs. SO isn't exactly an unbiased comparison since it's more programmer oriented (but it's also not exactly an academic haven...). But at least it provides some evidence to show that if noetsi wants to claim that SAS is king then he needs to bring some actual data to the table.
 

noetsi

Fortran must die
#27
There is data.
There is data.
There is data.
There is data.
There is data.
There is data.
http://r4stats.com/articles/popularity/
You cannot keep ignoring this and spewing this unfounded **** about SAS being more popular than R in general.
Are there specific pockets of industry where SAS is more popular than R? Almost certainly. It looks like you work in such a pocket. But to say that SAS is more popular than R in general flies in the face of the evidence.
there is data you prefer that may or may not get at systematic factors. The author I posted, which you ignored said the exact opposite. Nothing in that article convinced me they has done a reasonable sample of private or state organizations and based their findings on that. Using what was asked for in one job board as a way of analyzing anything is a joke.

I am quite sure you would not apply such an unsystematic methodology in any serious analysis.
 
#30
SAS v R.

the bottom line is that (according to the authors anyhow) R is better for cutting edge statistics (new methods not seen before) and SAS is easier to learn and more likely to be used in private organizations (aka companies). Although they do not say this outright it suggests that R is more used by academics.
Yes, SAS is easy to learn, when compared to R.
But SAS' syntax is weird, considered from perspective of a programmer.

I've no idea why SAS is soooo expensive, any person of average salary won't be able to afford it, the only way to get a copy is to get the free academic version (which became available only after SAS lost the legal case to WPS).
 

hlsmith

Omega Contributor
#31
Yes, I will echo again that certain sectors (e.g., loans, insurance) require transparency and consistency in their models. This traditionally has been met through the use of SAS. The product is too expensive and has to be getting hit with fewer relative users every year as academia is switching over to using open-source more and more. I will add that traditionally, users could create macros in SAS where the code was available and these "packages' would eventually be absorb into SAS. Though I would imagine now days people are more inclined to create these in R and get better authorship for them while hosting them on GitHub. Which I believe was purchased by MS for billions. I will also note that version control is an issue in R. Though most updates and dependency issues are not of major or substantive concern to some users. Though certain sectors may require them to host their own version of the package to ensure consistency.

Also, given the general and European push toward transparency in supposed black box models, new R packages are coming out (e.g., LIME, etc.). I would imagine these advances will also lead toward less dependence on SAS, who will have a lag in the incorporation of them into packages.

A big question may be in the business sector, how much they will expect their employees to know how to code in the future and move away from easier point clicking in SAS and toward programming. There are probably 100,000's of business folks using SAS, are they able and ready to switch over to programming. That may need to happen at the academic level to prep them.