Unfortuantly they are not error free. If you look into their robust regression (Proc Robustreg) techniques they have plenty of errors in the way they process their residual analysis. Their out of sample fitting routines (R-square hold-out in Time series for example) is computed entirely wrong. I have talked with them many times about their errors in both Base SAS as well as their Time Series Forecast System platforms as I have been one of their beta testers in the past. That is not to say they are not the best stat language out their (with S-plus a close second) but unless you have an API tie in, and a way to quickly process large Proc SQL queries, it makes it a very difficult program to fully automate. This doesn't even address the efficiency issues in large Marco curve fitting algorithims which these systems are usually based on. Maybe this can be solved with super heavy duty servers or something along those lines but even with a tricked out 64 bit computer it still runs into serious system bog down issues.
agreed. the only reason matlab has a strangle hold is because people coming out of school know it. that's slowly changing since more schools are using alternatives ...( python ipython/scipy/numpy).
I use MATLAB- not for trading, but for data mining at a global bank. It is my tool of choice. I've used SAS quite a bit, but avoid it if at all possible.
Numeric runs all 18B on SAS. I should have added that we are not doing any high frequency stuff with this...some intra day but no high freq.
If OP is already a SAS license-holder, hey, he can give it a go and let us know how it works out. If he isn't, I'm at a loss as to how/why he would focus on SAS as a prospective toolset.
Hmmm...any software vender that exposes their platform to windows through an api can potentially be used as a trading device in the windows environment. Sas, matlab, whatever....I really don't see the issue of argument here. If it is a systematic and potentially slow trade, more power to you. You will be the first person I know of doing this with sas though so enjoy being the canary.