Skip to content

How to choose appropriate linear model? (ANOVA)

5 messages · Peng Yu, Tal Galili, Rolf Turner +2 more

#
I'm wondering how to choose an appropriate linear model for a given
problem. I have been reading Applied Linear Regression Models by John
Neter, Michael H Kutner, William Wasserman and Christopher J.
Nachtsheim. I'm still not clear how to choose an appropriate linear
model.

For multi-factor ANOVA, shall I start with all the interaction terms
and do an F-test to see with interaction terms are not significant,
then do a linear regression on a model without the non-significant
iteration term?

Could somebody point me some good book or chapters on this topic?
#
On 19/11/2009, at 9:10 AM, Tal Galili wrote:

            
Alan Miller's book ``Subset Selection in Regression'' (Chapman and Hall,
1990) has some relevance.

	cheers,

		Rolf Turner

######################################################################
Attention:\ This e-mail message is privileged and confid...{{dropped:9}}
#
Hi,
On Nov 18, 2009, at 3:33 PM, Rolf Turner wrote:

            
You can also look into the "more recent" approaches, like penalized  
regression. Specifically I'm talking about the lasso or elasticnet.  
Look for the relevant papers by Trevor Hastie and  Tibshirani (you'll  
get them from their websites)

Lucky for you, the "glmnet" package is available for R, implements  
both the lasso and the elasticnet, and was written by these same people.

-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
#
On Nov 18, 2009, at 4:06 PM, Steve Lianoglou wrote:

            
Just for the record, Harrell's text cites, discusses and endorses  
penalized approaches. You can also read his more recent presentation  
at his website.