Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: Whish for Stata10, or earlier


From   "Earnhart, Benjamin J" <[email protected]>
To   <[email protected]>
Subject   st: RE: Whish for Stata10, or earlier
Date   Wed, 20 Dec 2006 04:52:44 -0600

dataset, and until you save it, it's pointless to refer to one dataset
or another -- "There can be only One."   If you're working with .do
files, then any worries about which dataset are moot, since you
deliberately open and close them as you go.  Regardless, the current
data is stored in the "main memory" or "on-chip memory."  With ODBC/SQL
capabilities being developed, it is able to treat datasets as tables
sometimes, and allow multi-user access as well as files stored outside
of main memory.  But in general, it follows the most intuitive approach
(at least most intuitive to me), of having one dataset open at a time,
that can only be accessed by that one user, stored in memory, on-chip,
until it is saved.  Fast. Simple.

It happens to be a very efficient way to do things (the same procedure
in SAS or SPSS may take much, much longer than running it in Stata), but
it has its limitations (viz. Stata/SE, and even under /SE, effectively a
700-800 MB limit with 32-bit Windows).  Under alternate (64-bit) OSs,
it's (IMHO) keeping up with what's needed.  But what you consider a
problem is (as far as I understand things) a major plus in Stata's favor
under most circumstances.  

If you're Amazon.com, you want a tyrantasaur for data management, doing
"enterprise-wide-data-warehousing" and following buzzwords that boil
down to generating crappy real-time reports for middle managers who
don't understand them anyway.  But if you're a mostly-stand-alone
scientist, a velociraptor is much more efficient.  Do it fast, and do it
right -- that's Stata.

Related to #2, the mere idea of a "data editor" window is anathema to
me: if you need to write a one-line command to change observation
#9876's value on v123 from 3 to 1, then at least it's potentially
documented.  Point-and-click modification of data is a recipe for
disaster.  That's even apart from the hit in efficiency, which would be
huge.

Just my 2 cents (which I think is only worth about 1.65 cents on the
other side of the pond).

Ben

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of roland
andersson
Sent: Wednesday, December 20, 2006 12:40 AM
To: [email protected]
Subject: st: Whish for Stata10, or earlier

I have two enoying problems with Stata9.

1) I do not know which data file I have open. It is common in other
programs to show the filename in the frame of the program.

2) I would like to execute commands with the dataeditor window open.
It would give a better "contact" with the data if I could easily
switch between the data and results windows.

When can we have these functions in Stata?

Roland Andersson
*
*   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/

*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index