Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

st: rescaling of covariates in logistic regression


From   Stephan Rudolfer <[email protected]>
To   "Stata list ([email protected])" <[email protected]>
Subject   st: rescaling of covariates in logistic regression
Date   Thu, 31 Mar 2011 19:42:44 +0000

I am working with a neural networks expert, who informs me that networks require the covariates to be rescaled linearly to the interval [-1,1] from -1 to 1.  He also suggested that I might carry out the same transformation prior to fitting a binary logistic regression model.

It is clear that the predicted values of the logistic regression model will remain unchanged by linear coding of the covariates.  My question is whether there are any other considerations I should take into account.

Many thanks in advance for any suggestions/comments,
Stephan
--
Dr Stephan M Rudolfer
Honorary Research Fellow in Biostatistics 
Chairman, Christian Higher Education Network in Manchester
Website: http://www.christian-he-network.org.uk/
Biostatistics, Health Methodology Research Group
School of Community Based Medicine
Jean McFarlane Building
University of Manchester
Oxford Road, Manchester M13 9PL.
Tel: +44 161 275 7688    Fax: +44 161 275 5205
Email: [email protected]
Web (Maths pre-2005): http://www.maths.man.ac.uk/DeptWeb/Homepages/smr/
Web (Biostatistics 2005 -): http://www.medicine.manchester.ac.uk/staff/StephanMRudolfer 


*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index