Re: st: Stata 13 wishlist

[Christopher Baum <kit.baum@bc.edu>]

<>
On Jan 3, 2013, at 2:33 AM, J wrote:

> Ah, well I do like the graph system and the graph editor but the
> ability for MATLAB (is it all caps?) to take a graph "as is" and
> generate code for it is really great. A lot of times I would edit a
> graph by hand to get the graph roughly looking the way I wanted and
> then save the code and tweak it. In short, I used it the way I use
> Stata's GUI for syntax in complex procedures such as -xtmelogit-,
> where there are tons of options and having the GUI is cool to lay it
> all out in front of you.

Stata's capabilities are quite similar. If a .gph file is saved (without using the 'asis' option), it contains the code which produced it, the name and timestamp of the datafile from which it was constructed, etc. If you have a .gph lying around,

. graph use whatever
. graph describe
. ret li

Note also that by accessing the serset of the graph, you can actually extract the data, as illustrated in ITSP, 10.5. E.g.,

. loc vn: serset varnames
. di "`vn'"
. serset sum varname, detail
. ret li

Of course, if you're using a proprietary 'canned' solution such as Adept, it is likely that you do not have access to the .gph files. As Nick said, that system has Stata embedded, but tries to hide it. It also, in a most un-Stataish way, is not cross-platform capable.

Kit

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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/