--- bumbuminc <[email protected]> wrote:
> I mean
>
> Badi Baltagi wrote 2002 in "Econometrics" Page 84: In the
> multivariate analyses,the variable wage per hour are a log function.
> With these semi-log regression, the coefficient can in an exponential
> equation [exp(b)-1] interpret the covariate over the dependent
> variable as an percental effect.
In the example I gave you earlier you will be able to make a statement
like the the hourly wage changes by a factor of 1.08 for a year
increase in education. This, obviously implies an 8% increase. In my
particular sub-sub-discipline people are used to interpreting odds
ratios, which have a very similar "factor change" interpretation, so I
won't bother with subtracting 1, and multiplying by 100. Whether that
works in your sub-sub-discipline is up to you.
-- Maarten
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room N515
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
