Re: st: RE: RE: The Future of Statistical Computing

[Diego Navarro <the.electric.me@gmail.com>]

The fidelity of users to Stata indeed mirrors that of Apple. The thing  
that always puts me off about R is that it doesn't really seem to be  
oriented around datasets -- rather, it comes across as a matrix  
programming language with a Common Lisp-like object system. Most of  
the graduate-level courses I took followed matrix-oriented econometric  
theory manuals and were actually accompanied by R practical exercises  
-- which was immensely useful (I cherish the nights spent trying out  
variations of bootstrapping algorithms) as a learning device -- but  
when I stepped out into the "real world", time constraints first  
dictated that I used the house standard (eViews) until I came across  
Stata. Now, Stata's language is exceedingly ad hoc, and it could  
withstand, with little actual changes in current syntax, several  
rounds of formalization until a formal calculus of Stata predicates  
could be constructed. The main reason for Matlab (and R) being the  
dominant languages for scientific computing, I feel, is that  
scientists are used to formulate their problems around matrix  
representations in order to prove theorems about them. The success of  
Stata is witness to the real-world need of "xBase/Clipper"-like  
constructs, but statisticians (specially econometricians) need to be  
able to prove theorems in a formal system including all such  
constructs. Mata is pragmatically necessary right now, but it's a bit  
of a step down from Stata's strengths.

Sorry if this comes across as a sequence of nonsequiturs. I don't have  
much time to write right now, so I might have erred on the side of  
concision.

--
Diego Navarro
(21) 2559-5620

“The first step is to measure what can be easily measured.  This is  
okay as far as it goes.  The second step is to disregard that which  
cannot be measured, or give it an arbitrary quantitative value.  This  
is artificial and misleading.  The third step is to presume that what  
cannot be measured really is not very important.  This is blindness.   
The fourth step is to say that what cannot be measured does not really  
exist.  This is suicide.” (Daniel Yankelovich, 1973)


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/