Dear all: I try to analyse a dataset which contain one binary response variable and serveral predict variables, but multiple colinear problem exists in my dataset, some paper suggest that logistic regression for principle components is suit for these noise data, but i only find R can done principle component regression using "pls" package, is there any package that can do the task i need - logistic regression based on principle components, if not, can anyone give some suggestion about how to use R to do my work. Thanks very much! best regards! wenkai _________________________________________________________________ [[elided Hotmail spam]]
logistic regression based on principle component analysis
4 messages · 江文恺, Steve Lianoglou, Kjetil Halvorsen +1 more
Hi,
On Thu, Jan 7, 2010 at 11:57 AM, ??? <biology0046 at hotmail.com> wrote:
Dear all: I try to analyse a dataset which contain one binary response variable and serveral predict variables, but multiple colinear problem exists in my dataset, some paper suggest that logistic regression for principle components is suit for these noise data, but i only find R can done principle component regression using "pls" package, is there any package that can do the task i need - logistic regression based on principle components, if not, can anyone give some suggestion about how to use R to do my work.
Is this any different than first doing PCA to do the dimensionality reduction (which presumably will take care of your colinearity), then doing the logistic regression on your reduced input space? If so: no package is really necessary, right? It's just a two-step solution you need to write up. -steve
Steve Lianoglou Graduate Student: Computational Systems Biology | Memorial Sloan-Kettering Cancer Center | Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact
for an alternative (lasso) approach, look at the packages (CRAN) grpreg, grplasso, glmnet, penalized and certainly some others. Kjetil B H On Thu, Jan 7, 2010 at 2:06 PM, Steve Lianoglou
<mailinglist.honeypot at gmail.com> wrote:
Hi, On Thu, Jan 7, 2010 at 11:57 AM, ??? <biology0046 at hotmail.com> wrote:
Dear all: I try to analyse a dataset which contain one binary response variable and serveral predict variables, but multiple colinear problem exists in my dataset, some paper suggest that logistic regression for principle components is suit for these noise data, but i only find R can done principle component regression using "pls" package, is there any package that can do the task i need - logistic regression based on principle components, if not, can anyone give some suggestion about how to use R to do my work.
Is this any different than first doing PCA to do the dimensionality reduction (which presumably will take care of your colinearity), then doing the logistic regression on your reduced input space? If so: no package is really necessary, right? It's just a two-step solution you need to write up. -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology ?| Memorial Sloan-Kettering Cancer Center ?| Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact
______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Frank Harrell, Jr shows you how to implement this in R, in his book, Regression Modeling Strategies. ~~~~~~~~~~~ Scott R Millis, PhD, ABPP (CN,CL,RP), CStat, CSci Professor & Director of Research Dept of Physical Medicine & Rehabilitation Dept of Emergency Medicine Wayne State University School of Medicine 261 Mack Blvd Detroit, MI 48201 Email: aa3379 at wayne.edu Email: srmillis at yahoo.com Tel: 313-993-8085 Fax: 313-966-7682
--- On Thu, 1/7/10, Kjetil Halvorsen <kjetilbrinchmannhalvorsen at gmail.com> wrote:
From: Kjetil Halvorsen <kjetilbrinchmannhalvorsen at gmail.com> Subject: Re: [R] logistic regression based on principle component analysis To: "Steve Lianoglou" <mailinglist.honeypot at gmail.com> Cc: r-help at r-project.org, "???" <biology0046 at hotmail.com> Date: Thursday, January 7, 2010, 12:27 PM for an alternative (lasso) approach, look at the packages (CRAN) grpreg, grplasso,? glmnet, penalized and certainly some others. Kjetil B H On Thu, Jan 7, 2010 at 2:06 PM, Steve Lianoglou <mailinglist.honeypot at gmail.com> wrote:
Hi, On Thu, Jan 7, 2010 at 11:57 AM, ??? <biology0046 at hotmail.com>
wrote:
Dear all: I try to analyse a dataset which contain one
binary response variable and serveral predict variables, but multiple colinear problem exists in my dataset, some paper suggest that logistic regression for principle components is suit for these noise data,
but i only find R can done principle component
regression using "pls" package,
is there any package that can do the task i need -
logistic regression based on principle components,
if not, can anyone give some suggestion about how
to use R to do my work.
Is this any different than first doing PCA to do the
dimensionality
reduction (which presumably will take care of your
colinearity), then
doing the logistic regression on your reduced input
space?
If so: no package is really necessary, right? It's
just a two-step
solution you need to write up. -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology ?| Memorial Sloan-Kettering Cancer Center ?| Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact
______________________________________________ R-help at r-project.org
mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained,
reproducible code.
______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.