s-news
[Top] [All Lists]

[S] parentheses in interaction terms

To: s-news@wubios.wustl.edu
Subject: [S] parentheses in interaction terms
From: "Isabelle Robichaud" <eir8049@umoncton.ca>
Date: Fri, 22 Jan 1999 19:01:52 +0800
Reply-to: eir8049@umoncton.ca
Sender: owner-s-news@wubios.wustl.edu
Dear fellow S-Plus users,

I am relatively new at S-Plus, which might explain my question, but 
I have come across a behavior that puzzles me.  Any 
insight/comments would be very appreciated.

I obtain different results whether I put parentheses or not around 
the terms in interactions.  For example, glm(RESP ~ X1:X2) gives 
different results from glm(RESP ~ (X1):(X2)).  In fact, the second 
model gives the same deviance and df as glm(RESP ~ X1*X2) 
does.  Is this normal?  (X1 is a continuous variable, X2 is an 
ordered factor.)

Sorry if the question is evident to everyone but me.  Thanks in 
advance!

Isabelle
p.s. I work with version 4.5 (professional, release 2) on a Pentium 
233, 32 M of RAM.

Isabelle Robichaud
Departement de biologie
Universite de Moncton
Moncton NB Canada E1A 3E9
e-mail: eir8049@umoncton.ca

-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu.  To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message:  unsubscribe s-news

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