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

st: boxtid & linearity


From   SR Millis <srmillis@mindspring.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: boxtid & linearity
Date   Thu, 17 Apr 2003 12:52:17 -0400

I've run a logistic regression model. I wanted to evaluate whether there was substantial nonlinarity in the logit, so I used the Box-Tidwell transformation and found that there was nonlinearity, as shown in the following output. However, when I followed up with mfracpol, I wasn't able to find any alternative tranformation. Any suggestion?

Thanks,
SR Millis

note: 0 failures and 16 successes completely determined.
ldfr | -.2405828 .0990139 -2.430 Nonlin. dev. 9.295 (P = 0.002)
p1 | 4.520587 2.083727 2.169
------------------------------------------------------------------------------
sdfr | .2828134 .1111305 2.545 Nonlin. dev. 5.292 (P = 0.021)
p1 | 3.944163 1.872817 2.106
------------------------------------------------------------------------------
hits | -.6063142 .1097356 -5.525 Nonlin. dev. 3.331 (P = 0.068)
p1 | -2.475662 2.074093 -1.194
------------------------------------------------------------------------------
Deviance: 117.923.




*
* 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