Maoyong Fan wrote:
> Is there any way to just put my regression results instead of every step of
> my program into the log file? That will make it easier to check results.
Try putting the regression command in a brief, simple "wrapper" program, along
with the -log using- and -log close- commands.
The log file will only have the regression results--it won't even contain the
regression command, itself.
I recommend adding a brief -display- command just before the regression command
to provide identifying information in the log file. You can even -display- the
regression command (in quotes) in order to identify it.
The approach is illustrated in the do-file below. You can also use -log off- /
-log on- in this approach with the necessary modifications to the program(s).
Joseph Coveney
-------------------------------------------------------------------------------
clear
set more off
sysuse auto
*
* data-manipulation steps of your program
* that you don't want to log go here
*
program define logem
version 8.2
capture log close
quietly log using my_log
display as text "This is the title of my regression (to identify it)"
display
regress `1' `2'
quietly log close
end
logem headroom gear_ratio
exit
--------------------------------------------------------------------------------
*
* 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/
