[R] glm

Murray Jorgensen maj at waikato.ac.nz
Thu Dec 21 05:23:50 CET 2000

Just as a matter of interest, how did this strange form for the LHS of
glms evolve? When my students ask I usually mutter something about a smooth
generalization to the multinomial case, but I've no idea if this was really
motivation for doing things this way.

Murray Jorgensen

At 11:42 AM 20-12-00 -0500, Robert Gentleman wrote:
>On Wed, Dec 20, 2000 at 05:01:44PM +0100, Oscar wrote:
>> It was my fault, as you have seen. Y matrix must contain successes and
failures, and not successes and trials.
>> Sorry for my ignorance and thanks a lot for your help.
>> Oscar
>  This is a very good argument for a more sophisticated modeling
>paradigm where the lefthand side must be a Binomial object. Filling
>this up correctly is often easier and more straight forward than
>trying to figure out what is wanted at fit time.
>| Robert Gentleman                 phone : (617) 632-5250                   |
>| Associate Professor              fax:   (617)  632-2444                   |
>| Department of Biostatistics      office: not yet                          |
>| Harvard School of Public Health  email: rgentlem at jimmy.dfci.harvard.edu   |
>r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
>Send "info", "help", or "[un]subscribe"
>(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
Murray Jorgensen,  Department of Statistics,
University of Waikato,  Hamilton,  New Zealand.  [maj at waikato.ac.nz]
"These data do not support the hypothesis"      Well . . .
The first one does, but the second and third don't, now the fourth . . .
[ In which I pose the question: is "data" really plural. ] 
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list