Message-ID: <5.1.0.14.2.20030605105306.01f894b0@mcmail.cis.mcmaster.ca>
Date: 2003-06-05T15:01:17Z
From: John Fox
Subject: coefficient of logistic regression
In-Reply-To: <3EDDE036.2010206@deprem.gov.tr>
Dear can y,
At 03:04 PM 6/4/2003 +0300, orkun wrote:
[previous messages deleted]
>Dear Mr. Fox
>
>thank you very much all.
>
>Because of related to your answer. I ask you directly if you don't mind
>I studied several ways after my email.
>I wonder whether pgeo<-predict.glm(glm.ob,type="terms")
>gives same result with probability value I asked before.
>I tried on it. But it gives "Error in rep(1/n,n) %*% model.matrix(object):
>non conformable
>arguments" .
I don't know why this doesn't work for you -- it works for me. I don't
think that this will give you what you want, however: setting type="terms"
produces the (centred) term-wise components of the fitted values on the
scale of the linear predictor (i.e., the logit scale).
I think that the responses that you got previously from Tom Blackwell and
from me answer your question.
>By the way , your teaching notes is available on the internet ?
>
They are, along with other course materials, at
<http://www.math.yorku.ca/SCS/spida/glm/>. Unfortunately, these workshops
were taught using SAS rather than R (not my choice).
John
-----------------------------------------------------
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario, Canada L8S 4M4
email: jfox at mcmaster.ca
phone: 905-525-9140x23604
web: www.socsci.mcmaster.ca/jfox