Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: logit: t-test for odds-ratio vs coefficient


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   stata list <statalist@hsphsun2.harvard.edu>
Subject   RE: st: logit: t-test for odds-ratio vs coefficient
Date   Mon, 10 Aug 2009 11:58:00 +0000 (GMT)

--- Kyle Caswell wrote:
> I have been using -ologit- with survey data (i.e., using svy) and am
> reporting only odds-ratios, not coefficients, for a write-up. When
> using the option "or" (for odds ratio) Stata produces a table with
> transformed standard errors and confidence intervals for the reported
> odds-ratios. However, the t-statistics and corresponding p-values are
> the same as in the output when the option "or" is not specified. In
> other words, I understand that the t-stats and P>|z| refer to the null
> hypothesis of whether a particular estimated *coefficient* is equal to
> zero.
> 
> Is there an option which produces t-statistics, and P>|z|, corresponding
> to the null hypothesis that the given odds-ratio displayed in the table
> is equal to one?

The coefficient is the log odds ratio, so a test that the log odds ratio is equal to zero is equivalent to a test that the odds ratio is equal to one. So, the reported t-values and p-values, are the values you want.

Hope this helps,
Maarten

-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index