Hi Dan
This is a subject of special interest to me. You will find the papers at
the link below very helpful.
http://www.logitresearch.com/Library_of_technical_papers
The last paper in particular you might find the following paper of
interest.
Although a lot of the material is based around the Correlated Component
Regression (CCR) methodology for linear and logistic modelling, which is
geared around high dimensional data (few cases/many correlated
predictors) there is also a lot of general background material on the
issue you raise in these papers. If you are interested we are running a
course, led by Dr Magidson, on the topic which will cover this issue in
depth using material not publically available on the eve of the Royal
Statistical Societies conference in Telford:
http://www.logitresearch.com/Training_and_Development
Regards
Gary
_______________
Gary Bennett
Logit Research
Tel: +44 (0)1732 465662
Fax: +44 (0)1732 465662
Mob: +44 (0)7795 342575
http://www.logitresearch.com/
Follow Gary on Twitter http://twitter.com/GaryLogit
Logit Research Limited, Unit 3, 7 Blighs Walk, Sevenoaks, Kent. TN13
1DB. Company Reg No. 5086604. VAT Reg No. 826 5147 25
-----Original Message-----
From: A UK-based worldwide e-mail broadcast system mailing list
[mailto:[log in to unmask]] On Behalf Of Dan Abner
Sent: 28 June 2012 16:31
To: [log in to unmask]
Subject: Multicollinearity affecting prediction XXXX
Hi everyone,
I am in need of some info concerning how (if at all) multicollinearity
in a generalized linear model affects prediction. I am quite familiar
with the impact on explanation, however, my current endeavor is purely
predictive.
Can anyone suggest some citations please?
Thanks!
Dan
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|