As SSC maintainer, I would recommend circulating this new version of
outreg as outreg2. I do not replace material on SSC without the
author's request unless (a) it is clearly disfunctional or (b) the
author has accepted the new person as the new maintainer and coauthor
of the routine. Since outreg works for some people some of the time,
it should stay on SSC. I would be glad to host -outreg2-, Roy's
modified version, on SSC.
Per my previous message, I would personally find such a routine of no
value unless it could handle LaTeX output, a simple transformation of
tab-delimited output. If Roy would like the outreg code for Gallup's
version 4.x, which does have LaTeX output options, he should contact
me privately. I used outreg with the LaTeX option for several years
in my own work, and for what I was doing it did a nice job -- but the
program required frequent tweaking, since it did not know how to
handle commands that I had written (understandably). If Roy doesn't
want to be bombarded with questions (cluttering up Statalist!) he
should figure out how to make outreg emulate estout and accept the
results of any estimation command, using what has been left behind to
determine what can be presented for that particular command.
Otherwise outreg(2) will always be a poor second to estout in its
capabilities to get the job done without further programming.
Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html
On Sep 15, 2005, at 2:33 AM, Roy wrote:
A number of interesting points have been brought up by a number of
people.
Richard has suggested naming the updated version as -outreg2-. This
seems
like a good idea. If John Gallup ever comes back, as Nick has
mentioned, he
can pick up where he has left off.
I think I understand the program well enough to make small
adjustments as
necessary. I have worked with parsing and tokens before. I am not
offering
to do a major overhaul at this point.
If you are asking me if I will hold everyone's hand who comes my
way, I have
not done it before and I probably will not do it like Nick has. I'm
not a
nice guy like Nick. But I will offer enough support to keep the
program
going in the near future.
The primary reason I have offered to update -outreg- is because no
one else
has offered to do so, despite the obvious need and the frequent
requests to
do so during the past couple of years.
It might be the Bambi speaking here, but I always thought -outreg-
is a nice
piece of work. Just like Steve, I like to use it for its ease of
use. It's a
write-and-forget command that I don't have to worry about.
One drawback of -estout- (it's a well-written program, by the way) is
requiring the user to actively manage the estimates. The added
layer of
control let you doing things, but it comes at a cost. The users are
required
to name each estimate, and then call them up under the -estout-
command.
I plan on further testing -outreg2- this weekend. If you are
interested in
helping me with testing (with your own data), please contact me
privately
and I will email the beta version to you next week.
*
* 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/
