[R] Anova of a nls object

[Bill.Venables at csiro.au]

You don't, because the operation is not well defined.

Using anova( ) on a single linear model object will fit a sequential
sequence of sub-models got by adding the terms into the linear model, as
they appear in the model formula, one by one.

In non-linear regression there is no unique definition of sub-model.
You may have a non-linear regression consisting of a sum of terms, but
if that's the way you wish to define the sequential decomposition, you
have to do so explicitly and fit the models by hand.

(Some people have doubts about the wisdom of allowing anova() even on a
single linear model object, since the sequential of sub-models used is
arbitrary even there.  I think this view has a lot going for it.) 


Bill Venables
CSIRO Laboratories
PO Box 120, Cleveland, 4163
AUSTRALIA
Office Phone (email preferred): +61 7 3826 7251
Fax (if absolutely necessary):  +61 7 3826 7304
Mobile:                         +61 4 8819 4402
Home Phone:                     +61 7 3286 7700
mailto:Bill.Venables at csiro.au
http://www.cmis.csiro.au/bill.venables/ 

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]
On Behalf Of Eric Ferreira
Sent: Thursday, 20 March 2008 4:21 AM
To: r-help at r-project.org
Subject: [R] Anova of a nls object

Dear useRs,

How can I perform the analysis of variance of just one nls object?

Regards


-- 
Eric B Ferreira
Exact Sciences Department
Federal University of Lavras
Minas Gerais - Brazil

	[[alternative HTML version deleted]]

______________________________________________
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.