Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

re: st: list x matrix


From   Kit Baum <Baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   re: st: list x matrix
Date   Sun, 28 Mar 2010 11:04:22 -0400

<>
Maarten said, in response to my Mata posting,

If we take Richard's question literaly, then this if probably
the best answer. However, I would argue that the fast 
majority of users should *not* use it, this includes users who
program in Stata.

I wholeheartedly agree. If a tested and supported command in Stata will do the job, users should take the time to seek 
it out and use it. The fact that one _can_ do any matrix computation quite easily in Stata/Mata does not mean that it is a good
idea for users to 'roll their own' solution. The recent interchange on this list on computational issues with -margins- is
a good case in point. The way that Stata calculates something as simple as  a linear regression is much more
numerically stable that a brute-force computation of (X'X)^{-1} (X'y). For pedagogical purposes, using the latter is 
fine. For serious research that could involve ill-conditioned regressor matrices, it is decidedly not recommended.

Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html


*
*   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/


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