s-news
[Top] [All Lists]

Use of step.gam

To: s-news@lists.biostat.wustl.edu
Subject: Use of step.gam
From: Brian Newquist <BNewquist@constellagroup.com>
Date: Tue, 15 Feb 2005 18:20:16 -0500

Is anyone familiar with the use of the step.gam function?  I am trying to find the best value for the span parameter for use in the gam function.  The S-Plus help documentation suggests that step.gam will return the best model (in terms of AIC criteria) from among those that you have defined in the scope argument.  

 

The example below illustrates what occurred when I used this function.  Does anyone know where I went wrong?  Any help will be greatfully appreciated. 

 

fit.gam <- gam( OUTVAR ~ LOGVAR + VAR1 + VAR2 + VAR3, binomial, data = "">

step.gam(fit.gam, scope=list(

   "LOGVAR" = ~ lo(LOGVAR, span=0.1) + lo(LOGVAR, span = 0.15) + lo(LOGVAR, span = 0.20)))

 

The following message appears after running step.gam :

Problem in untangle.scope(object$terms, scope): regimen 1 does not appear in the initial model

 

 

Brian C. Newquist

Research Statistician

Constella Health Sciences

2605 Meridian Parkway, Suite 200

Durham, NC 27713

Tel:  (919) 313-7588

Fax: (919) 544-7507

bnewquist@constellagroup.com

www.constellagroup.com

 

 

 

<Prev in Thread] Current Thread [Next in Thread>