There was a thread on the use of -nl- in these cases which obviates the need for a new var. Also note that -mfx- can calculate the marginal impact after an estimation command as d(y)/d(lnx)...
HTH
Martin
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Jan Sauermann
Sent: Tuesday, August 19, 2008 10:49 AM
To: [email protected]
Subject: st: use of logarithmised values in regressions (w/o generating new variables)
My second question this morning: I'm using logarithmised variables in
regressions. This can be done by generating a new variable:
gen LOGVAR = log(var)
reg outcome LOGVAR
Is there also a shorter way (without creating a new variable)? I.e.
something like this:
reg outcome log(var)
Thanks in advance,
Jan
--
Jan Sauermann (<[email protected]>)
Research Centre for Education and the Labour Market (ROA) / Maastricht
University
P.O. Box 616 | NL - 6200 MD Maastricht
office: +31.43.3883801 | mobile: +31.655853921
<http://www.roa.unimaas.nl>
*
* 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/
