It is not that surprising at all.
To clarify for a broad audience:
Many articles in the Stata Journal do not
have associated software, even if they
are largely or entirely about using Stata
commands.
Associated software is usually of the
form of .ado files and the associated
.hlp files. An important principle here
is that software published via the Stata
Journal should as far as possible be
comprehensible without further documentation
to those who do not have access to the
Stata Journal itself. Naturally the whole
point about a help file is to provide
the minimal documentation required to
understand the associated programs (and
sometimes more).
Incidentally, the preferred way
to download such files for use with
Stata should always be through Stata,
starting with a -search-. Stata knows
about its own .hlp files (unlike many
others Windows applications in particular)
and it also has sound ideas about where
to install them. (That said, there may
be firewall or other reasons why you
are obliged to download through the
browser.)
I can see the point about the code
in Allen McDowell's article. Someone
might want to start from that and modify
to suit their circumstances. However,
Allen's explanations are not phrased
in terms of .do files. If the Stata
Journal were to start publishing
code fragments as .do files, there
would be disadvantages:
(1) possibly lots of little .do files
that no one really wants
(2) undocumented files, and at least
some one expecting or demanding support
for them.
I don't think we are dogmatic about this.
We'd listen to a case for including
(substantial) code fragments as ancillary
files downloadable from the Stata Journal
website. But on the whole, authors' typical
do files would not be much more portable
or useful to others than users' typical do files.
Nick
[email protected]
Kit Baum
> Eric points out that the link
> http://www.stata-journal.com/software/sj4-4/st0079/
> does not work. Surprisingly, this article does NOT have any
> associated
> software. It would certainly be nice if at least the do-files
> quoted in the article were provided in machine readable form!
>
> You can examine what is available by looking at stata.toc in
> http://www.stata-journal.com/software/sj4-4
> which indeed shows no st0079.
>
> Just goes to show how useful it is for you (or your
> university library)
> to spring for a subscription to that excellent publication, the Stata
> Journal!
*
* 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/
