st: Re: re: do-file hygiene

["Martin Weiss" <martin.weiss1@gmx.de>]

Maybe a little unclear on my part: I tend to frolic in between runs of 
do-files to try out new things, and that is where the -cmdlog- is most 
helpful. Which is why you might want to start the -cmdlog- from your 
profile.do...

HTH
Martin
_______________________
----- Original Message ----- 
From: "Kit Baum" <kitbaum@mac.com>
To: <statalist@hsphsun2.harvard.edu>
Sent: Saturday, September 27, 2008 10:07 PM
Subject: st: re: do-file hygiene



> <>
> Martin said
>
> What I have learned, though, is that you must make sure you still 
> understand
> what you did half a year ago when that report comes back where the 
> referee
> says that you should really have included another year in your panel.  A 
> good
> way to do that is to have your do-files instruct Stata to keep both
> a-log-file, recording all screen output, and a command log. From my
> experience, many users do not know about -cmdlog-s, and if they do, they
> misunderstand the mechanism that allows simultaneous -log-ging
> and -cmdlog-ging. The trick is that there is a handle (the -name-  option)
> which is distinct from the -filename-...
>
>
>
> I fail to see what use a command log would be if created by running a 
> do-file. The commands therein will presumably be exactly those of the 
> do-file (unless you turn command logging off and on for some reason).
>
> The philosophy expressed in various contexts by StataCorp personnel 
> suggest creating a set of hierarchical do-files, similar to that used  in 
> producing a certification script (see Gould, 
> http://ideas.repec.org/a/tsj/stataj/v1y2001i1p29-50.html which is freely 
> downloadable). That is, your project (or  dissertation, or whatever) 
> master do-file calls do-files A1, A2, B1,  B2, B3, C1, ...  which may in 
> turn call subsidiary do-files. When you  execute the master file it runs 
> all of the do-files associated with  the project, each of which generates 
> a standard log file. This  approach both documents every step that you 
> take, from the first  access of the raw data to the final tables, charts, 
> estimation output,  etc. but the hierarchical arrangement means that you 
> can document the  order in which all of the do-files are run. If a new 
> processing step  is needed somewhere in the middle of the process that 
> do-file can just  be inserted in the sequence.
>
> IMHO command logs are useful in interactive use of Stata, but as I  would 
> always rely on a do-file to do anything serious, I don't see why  I would 
> want to have both foobar.do and foobar.txt, with foobar.txt  containing 
> the commands in foobar.do.
>
> Kit Baum, Boston College Economics and DIW Berlin
> http://ideas.repec.org/e/pba1.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/

*
*   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/