# st: AW: Creating a variable taking the value of elasticity

 From "Martin Weiss" To Subject st: AW: Creating a variable taking the value of elasticity Date Tue, 8 Sep 2009 15:58:23 +0200

```<>

Here is an example, although storage in a variable is not exactly
recommended. Note the last line which shows you how to get the correct name
for the coefficient...

*************
sysuse auto, clear

gen logmpg=ln(mpg)
gen loglength=log(len)
gen logturn=log(turn)

reg logmpg loglength logturn
*store in variable
gen mycoeff=_b[loglength]

l mycoeff in 1/10, noo

reg, coeflegend
*************

HTH
Martin

-----Ursprüngliche Nachricht-----
Von: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Dorothy Bridges
Gesendet: Dienstag, 8. September 2009 15:55
An: statalist@hsphsun2.harvard.edu
Betreff: st: Creating a variable taking the value of elasticity

Dear Statalist,

I have a simple dataset containing two variables, x and y.  I regress
ln(y) on ln(x) and observe the coefficient on x as elasticity.  I then
want to create a variable taking the value of this coefficient.  How
would I do this?  Something like local elas=e(b); gen elas=`elas',
except e(b) contains all the coefficients, and I just want one.

A related question: when I regress y on x and then input "mfx, eyex,"
the ey/ex value is not the same as the coefficient I obtain using
regress ln(y) ln(x).  Why?  Also: how would I extract the ey/ex value
from the mfx output in order to create a variable taking that value?

Thank you,
Dorothy

*
*   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/
```