Re: [R] (baseline) logistic regression + gof functions?

From: Wim Bertels <wim.bertels_at_khleuven.be>
Date: Fri, 06 Jun 2008 11:25:22 +0200

Thanks for the quick reply David,
so far this sums up to:

# logistic on binary data

  lrm combined with resid(model,'gof')

# logistic on raw binary data

  glm with gof using anova.glm
(i think that anova.glm only makes sence on grouped binary data, not on the raw binary data..) (so what is the gof for raw binary data and glm?)

# baseline logit on raw multicategory data
  multinom with no gof-test
(the only reference i found was *by GOEMAN* Jelle J. ; LE CESSIE Saskia eg https://openaccess.leidenuniv.nl/bitst*r*eam/1887/4324/22/04.pdf but there is no R implementation?)

(# also, what if the data are not binary and more than one predictor in the model?)

# what if the grouped data are very unbalanced and might have a lot of empty cell counts?

Hints? Suggestions? Other functions that might help?

eg:
by grouped data i mean (example for binomial):

    Var1 (with 3 levels)
    L1 L2 L3
T 1 1 2
F 0 2 0

by raw data i mean:
eg
Bin Var1

T      L1
T      L2
T      L3
T      L3
F      L2
F      L2

mvg,
Wim



R-help_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Fri 06 Jun 2008 - 12:08:44 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Fri 06 Jun 2008 - 12:30:38 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive