s-news
[Top] [All Lists]

glm warning message

To: s-news@lists.biostat.wustl.edu
Subject: glm warning message
From: Mike Trianni <mstdfw@yahoo.com>
Date: Tue, 7 Nov 2006 16:38:49 -0800 (PST)
Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:Received:Date:From:Subject:To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=PNdAi99hlSYqL1DxcLzMPh8FwpJHH0PmPi5Bm3Aijr0TacD1E2tlanjDfPPJ7OdrIcuc+bP+XrcOT6MVpk2zXpIe7MXvwd7oXutbvnpug/+ALu77EWaSSRM/6RS58QlJLwP1z2or27fclAI3r7BGSAgFKPXIIPXPUP3Z8tHAISs= ;
I have the following data frame, XX;
 NNN   OOO
 1  11   0.0
 2  11   0.0
 3  11   0.0
 4  12   0.0
 5  12   0.4
 6  13   0.5
 7  13   0.6
 8  13   0.7
 9  13   0.8
10  14  0.9
11  14  1.0
12  14  1.0
13  14  1.0
14  14  1.0
When I perform the following glm;
 
glm(OOO~NNN, family=binomial, data="">
 
in S-plus 6.2 there is no 'Warning message', but in S-plus 7.0 I get the following 'Warning message';
 
Warning messages:
  non-integer number of successes in a binomial glm in: glm.fitter(x = X, y = Y, w = w, start = start, offset = offset, family = family, maxit = control$maxit, epsilon =
 control$epsilon, trace = control$trace, null.dev = T)
Why no Warning meesage in 6.2?
 
Thanks for any insight
 
Mike
<Prev in Thread] Current Thread [Next in Thread>
  • glm warning message, Mike Trianni <=