It worked for Microsoft. Why mess with a winning strategy?

Because Microsoft is going to be doing the same thing, but better, as part of the ongoing roll-out of Microsoft Business Intelligence. [1]

Picking a symmetrical fight against Microsoft is mad-crazy.

[1] http://www.microsoft.com/bi/

It worked for Microsoft. Why mess with a winning strategy?

Thank you for commenting, but here and at other blogs. It is refreshing to see that SAS is paying attention to blogs!

From Anne Milley’s blog:

My remark reflects a key difference between R and SAS, that of support, reliability, and validation. Customers value SAS for many things, including our extensive testing, documentation, 24/7 support, and training. In contrast, the quality of proliferating R packages is varied and uneven, especially in complex analytical modules. Mistakes in these packages can lead to misleading results, even for experienced users.

First, SAS and I applaud the innovative contributions and passion of the R community, and users who apply R to solve problems. In a very real sense, we are grateful for R, as it provides a freely available venue for bleeding-edge and experimental data analysis methods, and underscores the increasing importance of advanced analytical and graphical methods in this age of massive data volumes.

strikes me as deceptive. The issue is not ‘bleeding edge’ or ‘experimental’ data analysis — the issue is correct data analysis. Both SAS and R have modules available from third-parties, and both include a core package.

SAS’s apparently strategy, of spreading fear, uncertainty, and doubt, is distressing.

In closing, I need to thank overdetermined [1], who pointed out me that SPSS now features optional integration with R. [2] Will SAS be doing the same soon?

[1] http://overdetermined.net/site/content/new-york-times-article-r
[2] http://www.spss.com/be/spss16/whats_new_base.htm

Well said!

The SAS representative may want to inquire about how HP-UX, AIX, Solaris, etc, market share is doing against Linux.

Searching the Harvard Business School website:

Excel: 1930 results [1]
SAS: 624 results [2]
SPSS: 2 results [3]
“R Project”: 0 results [4]

I have heard that SAS has a real advantage over R when it comes to very large data sets (something like more than a million observations). Certainly for projects that run into those considerations, it may be best ot use SAS or another proprietary piece of software.

However, considering the money SAS and SPSS get from college license fees, where students don’t do much more than regression, ANOVA, etc, I can’t see why the school would pay license fees for about $100/student/year (or force the students to do likewise) when the full R suite is available for free. Considering that most programs in the social sciences force every student to learn and use statistics,

At my university, the pure statistics program has completely moved to R, because R exposes the underlying matrix algebra that powers most of statistics. The applied stats program is beginning to move in that direction.

Even the infamous “Value at Risk” equation is available in R [5], for those financial wizards out there!

[1] http://www.google.com/search?hl=en&q=excel+site%3Ahbsp.harvard.edu
[2] http://www.google.com/search?hl=en&q=sas+site%3Ahbsp.harvard.edu
[3] http://www.google.com/search?hl=en&q=spss+site%3Ahbsp.harvard.edu
[4] http://www.google.com/search?hl=en&q=“r+project”+site%3Ahbsp.harvard.edu
[5] http://braverock.com/brian/R/PerformanceAnalytics/html/VaR.CornishFisher.html

I had the same eerie version of deja vu. It reminded me of what Michael Jordan used to say: “get out of my way and into my poster”. SAS must have missed out on the last decade even though Red Hat’s just down the street.

Makes you wonder what would happen if they taught R to MBAs. They’d probably produce a financial crisis that would dwarf the current one. R may be a threat to future financial stability.