Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: STATA Vs. SAS


From   Bill Rising <brising@louisville.edu>
To   "Stata Listserve" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: STATA Vs. SAS
Date   Wed, 27 Nov 2002 16:03:00 -0500

On 11/27/02 15:28, Nick Cox wrote

>Richard Herrell
>> 
>> On Wed, 27 Nov 2002, Nick Winter wrote:
>> 
>> > > -----Original Message-----
[snip...]
>
>I am still not clear whether -x- and -y- are variables, 
>and whether this all -- in Some Alternative
>Software -- is tacitly looping over 
>observations. (In my case, please assume 
>complete ignorance of SAS.) 
>
>I will also add that I don't find often myself 
>wanting to use completely different 
>commands for different subsets of a variable. 
>Alternatives could depend on what those 
>commands are. 

I've found one situation where this would be nice: when reading in really 
nasty data files. Two places where this happened in the past to me were:

Some really badly planned web forms, where I had to translate the 
resulting very messy, non-delimited emails. From what I heard (from a guy 
who had worked at SAS), SAS would be good for this, because of the 
ability to evaluate each incoming line and process it differently 
depending on its contents. I tried fixing things using lisp, ended up 
learning Perl and later wished that I had learned Python first.

Some badly made datasets from a weather data repository where it would 
have been nice to bring in the data line by line.

Now, while I sometimes wish that Stata could do this type of 
manipulation, I generally prefer to have Stata do what it does well 
(statistical analysis), and use other tools for complicated problems 
which Stata does not do well.

The same fellow who told me about how well SAS would have fit my needs 
also said that inside the SAS corp, SAS had to be called SAS (rather than 
its old name Statistical Analysis System), because the SAS folks regarded 
it as a data management package far more than a statistical package.

As for speed - in the distant past (oh, say, back in the mid 90s), I 
recall writing Stata code which would do in 15 hours what SAS would do in 
4 days. Of course - I suspect that the SAS programmers whose code I was 
duplication were not particularly attuned to being efficient. Things 
could be very different now.

Bill
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index