Hello,
I'm new to the list, hello everybody! :-)
I have a question. I looked carefully through the documentation and
googled, and found no answer, so I'm posting it here.
Let's say I have a data frame with lots of columns (about 300),
which
are factors. Many columns-factors have only one level, for example
all
are "Yes" or all are "No".
If I try to do a regression with that, I get an error message, that
contrasts can be used only when number of factors is at least 2.
glm(mortality ~ ., family = binomial, data = ecdb_PROC88)
Error in "contrasts<-"(`*tmp*`, value = "contr.treatment") :
contrasts can be applied only to factors with 2 or more
levels
Of course, I could not include them in the model formula, but I
don't
feel like typing 150 descriptive variables.
I would like to remove the one-level columns autimatically from the
data
frame. Is it possible?
Regards,
Maciej