Statalist


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

st: RE: Logistic regression with a continuous variable


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Logistic regression with a continuous variable
Date   Sun, 8 Mar 2009 18:31:58 -0000

In addition, this won't work unless all values of X are positive. 

Nick 
n.j.cox@durham.ac.uk 

SR Millis

To assess nonlinearity in the logit, you can use the Box-Tidwell transformation.  This involves adding the term, (X)ln(X), to the equation.  If the coefficient for this variables is statistically significant, there is evidence of nonlinearity in the relationship between logit(Y) and X.  However, this procedure isn't sensitive to small departures from linearity. In addition, this procedure won't tell you the precise form of the nonlinearity. 
 

*
*   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