 |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Re: log only results of r-class commands
Looks like the best strategy would be to try both your programs and see
which best suits the purpose. You're correct, I'd prefer several large
tables to many small tables, but this is for an appendix, where the
requirement is to show the stats, so I'm not picky on format. I'll
hand-craft the tables that appear in the body of the report.
I'm not sure from what you say if your program based on -outreg2- makes a
call to
-outreg2- and thus wouldn't work with the current version of -outreg2- or if
it emulates some features of -outreg2- and so in its current version would
not have all the bells and whistles of the latest -outreg2-.
If the latter, I'd appreciate the opportunity to try both solutions, either
through an ssc download (although at this hour repac.org isn't responding to
Stata calls from my location) or through an attachment to my email address
in the header.
In either case, I'll post to Statalist the results of my trials.
Steve
From Roy Wada <[email protected]>
To <[email protected]>
Subject st: Re: log only results of r-class commands
Date Fri, 23 May 2008 16:55:01 -0700
The descriptive analysis extensively uses -fsum- (fsum.ado Full
SUMmary program v 2.4.0 fw 2may05), contributed by Fred Wolfe,
as it prints variable labels instead of variable names.
I would like to preserve the comments in the log, since they identify
analyses, but could live without that feature. Far easier to add a few
table headers to table groups than to edit out hundreds of lines of
command code.
What you want is to clean up log files or produce tables. I have programs
for both, but it usually comes out to be the same thing.
I have a program based on -outreg2- that can produce tables of
summary statistics. It will handle labels, titles, notes, blah, blah.
It is byable, but only column-wise. The problem is that it would
have to be upgraded to the newer version of -outreg2- that is
more flexible. Like -outreg2-, this method can produce one huge
table or many small table files. You might not like many small
tables.
The other program is designed to chew up log files in text format.
It is capable of digesting numbers and dumping them into tab-delimited
format (xml, dta). A log file is much easier to clean after being imported
as a data file, because you would say something like: drop if v0=="/*".
The problem is that tab-limited files may not be what you want, which in
case you would also import un-tabbed file, and export that column after
you have finished dropping what you didn't want using the information
from tab-limited columns.
Roy
*
* 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/