Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: Replicating -margins,eydx()- with interactions
From
Duha Altindag <[email protected]>
To
statalist <[email protected]>
Subject
st: Replicating -margins,eydx()- with interactions
Date
Fri, 3 Jun 2011 14:00:36 -0500
Dear Stata listers:
I am having trouble replicating semi-elasticity output of
-margins,eydx()- when I use interactions.
Below is my code:
****
webuse auto,clear
g log_mpg=ln(mpg)
reg price i.foreign##c.log_mpg
margins foreign,eydx(log_mpg) atmeans
su price
di "elasticity at foreign=0: " _b[c.log_mpg]/r(mean)
di "elasticity at foreign=1: " (_b[c.log_mpg]+_b[1.foreign#c.log_mpg])/r(mean)
****
I know eydx calculates (dydx)*(1/y).
I am thinking the dydx (coefficients) that -margins- use to calculate
eydx must be same as those I am using. Then, the difference can be
only due to the denominator.
Should I be using a different mean instead of r(mean)?
Thanks for your consideration.
--
Duha T. Altindag
*
* 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/