: [R] fitting in logistic model
Thank you very much for your answer.
Mark hit the point of my query. Now we need somebody who knows how R
computes the fitting values, and why it does not use the inverse link...
In my humble opinion I think that R uses a kind of interpolation, using
some standard
]
Inviato: 30 June 2009 19:54
A: Marc Schwartz
Cc: r-help@r-project.org; Renzi Fabrizio
Oggetto: Re: [R] fitting in logistic model
On 30-Jun-09 17:41:20, Marc Schwartz wrote:
On Jun 30, 2009, at 10:44 AM, Ted Harding wrote:
On 30-Jun-09 14:52:20, Marc Schwartz wrote:
On Jun 30, 2009, at 4:54 AM
I would like to know how R computes the probability of an event in a
logistic model (P(y=1)) from the score s, linear combination of x and
beta.
I noticed that there are differences (small, less than e-16) between the
fitting values automatically computed in the glm procedure by R, and the
On Jun 30, 2009, at 4:54 AM, Renzi Fabrizio wrote:
I would like to know how R computes the probability of an event in a
logistic model (P(y=1)) from the score s, linear combination of x and
beta.
I noticed that there are differences (small, less than e-16) between
the
fitting values
On 30-Jun-09 14:52:20, Marc Schwartz wrote:
On Jun 30, 2009, at 4:54 AM, Renzi Fabrizio wrote:
I would like to know how R computes the probability of an event
in a logistic model (P(y=1)) from the score s, linear combination
of x and beta.
I noticed that there are differences (small, less
On 30-Jun-09 17:41:20, Marc Schwartz wrote:
On Jun 30, 2009, at 10:44 AM, Ted Harding wrote:
On 30-Jun-09 14:52:20, Marc Schwartz wrote:
On Jun 30, 2009, at 4:54 AM, Renzi Fabrizio wrote:
I would like to know how R computes the probability of an event
in a logistic model (P(y=1)) from
On Jun 30, 2009, at 10:44 AM, Ted Harding wrote:
On 30-Jun-09 14:52:20, Marc Schwartz wrote:
On Jun 30, 2009, at 4:54 AM, Renzi Fabrizio wrote:
I would like to know how R computes the probability of an event
in a logistic model (P(y=1)) from the score s, linear combination
of x and beta.
I
On Jun 30, 2009, at 12:54 PM, Ted Harding wrote:
On 30-Jun-09 17:41:20, Marc Schwartz wrote:
On Jun 30, 2009, at 10:44 AM, Ted Harding wrote:
On 30-Jun-09 14:52:20, Marc Schwartz wrote:
On Jun 30, 2009, at 4:54 AM, Renzi Fabrizio wrote:
I would like to know how R computes the
Snippage
On Jun 30, 2009, at 2:12 PM, Marc Schwartz wrote:
Wellbarring a correction from someone with more low level
insight, I would tend to suspect that the difference in our findings
may be attributed to some interaction of hardware, OS, compiler and
BLAS, perhaps weighted more to
9 matches
Mail list logo
