Hello All,
I am learning mixed effects logistic regression (lmer in lme4), and I am
having problems deciphering the goodness of fit function
plot.logistic.fit.fnc (languageR package). The only information I can find
is that this function plots observed proportions against mean predicted
probabilities for some binning of the data.
I have explored this function with several datasets, and have had some
problems. The resultant plots (for example from the bacteria dataset in
MASS) don't seem to match the predicted probabilities. For example, a given
dataset will have a maximum predicted probability of 0.7, but the plot will
show a point at 0.88. I have one dataset in which the "cut" method results
in negative observed proportions. So, I'm not sure that I really understand
how these mean predicted probabilities are calculated. Can somebody please
explain this or suggest a reference?
Thank you,
Leslie
[[alternative HTML version deleted]]
______________________________________________
R-help@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.
