I have little experience using -fracpoly- and I am seeking advice.
I have a variable AGE and a variable CAGE which is the centered version
of AGE and I want to find out whether in a binary logistic reg model I
should model age as linear or noinlinear in the logit. When I estimate
a fractional polynomial regression using AGE, the findings indicate I
should treat age as linear in the logit. When I use CAGE, the findings
indicate I should use a one-term polynomial transformation and model
CAGE as nonlinear.
Should I use the uncentered version in -fracpoly- or the centered
version?
Thanks in advance.
*
* 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/
