Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: logistic and logit and test


From   David Airey <david.airey@vanderbilt.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: logistic and logit and test
Date   Tue, 13 Apr 2004 18:34:12 -0500

I'm more used to continuous y's so this is probably a beginner's question.

"Logistic" in Stata produces an odds ratio estimate, while "logit" produces a coefficient. The standard interpretation of the odds ratio estimate is it is the number by which the odds favoring y = 1 is multiplied, with each 1-unit increase in the x variable, with any covariates held constant. The logit coefficient describes the change of the logit or log odds of y for each 1-unit change in x, with any covariates held constant. If the logit coefficient were say,

.6280264

this means that the logit or log odds of y increases ~.63 for every unit of x. Equivalently, we could say each additional unit of x multiplied the odds of y = 1 by

e^.6280264 = 1.87

All this has made me wonder whether the OR test is testing whether the OR is different from "1", whereas the test for the logit coefficient really is testing the coefficient being different from 0? Since exp(0) = 1, this suggests as much. But my question is that use of "test x" after either command produces the statement that x = 0 is the test.

Why am I confused?

-Dave

*
* 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/




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