Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Variables in logistic regression


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Variables in logistic regression
Date   Wed, 11 Jun 2008 14:27:03 +0100 (BST)

--- Alex Asiimwe <Alex.Asiimwe@port.ac.uk> wrote:
> Assuming I have age and gender, then I transform age to age*age to
> normalise it. From here, I want logistic regression model to predit
> an outcome- lets say mortality. Will I be wrong to use age, age*age
> and gender all at the same time in the model? If so, what are the
> implications? e.g logistic outcome age age*age gender; In otherwords
> is acceptable to use a variable and its transform in MLR (step wise)?

There is no problem to add age and age squared (age*age) to the same
model. What it means is that you think that the effect of age on the
log odds is non-linear, which in many cases is plausible especialy if
your age variable covers a large range. 

However:
- I would not call age*age normalizing
- there are many other ways of allowing for non-linear effects, e.g.
splines, see -help mkspline-, or fractional polynomials, see 
-help fracpoly-.
- stepwise is not acceptable, see:
http://www.stata.com/support/faqs/stat/stepwise.html

-- Maarten

-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      __________________________________________________________
Sent from Yahoo! Mail.
A Smarter Email http://uk.docs.yahoo.com/nowyoucan.html
*
*   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