If you don't want to generate a new variable you can do this is by
using lincom after your estimation command.
e.g.,
. sysuse auto.dta
. logistic foreign price
Logit estimates Number of obs = 74
LR chi2(1) = 0.17
Prob > chi2 = 0.6784
Log likelihood = -44.94724 Pseudo R2 = 0.0019
------------------------------------------------------------------------------
foreign | Odds Ratio Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
price | 1.000035 .0000844 0.42 0.676 .9998699 1.000201
------------------------------------------------------------------------------
. lincom 1000*price
( 1) 1000.0 price = 0.0
------------------------------------------------------------------------------
foreign | Odds Ratio Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | 1.03591 .0873912 0.42 0.676 .8780379 1.222168
------------------------------------------------------------------------------
hope this helps, Tim
On Thu, May 22, 2008 at 3:46 PM, <[email protected]> wrote:
> Dear Statalist
>
> I am performing a logistic regression using xtlogit. Some of my predictors
> are measured in days of treatment. Is there a way to recalculate the OR for
> a different unit of days, i.e., 30 days of treatment? Or would I have to
> transform the variable (e.g., dividing the per day variable by 30)?
>
> Thanks
>
> Barth
>
*
* 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/
