Dear All,
I am playing with different models/packages (random forest, logistic
regression, gbm etc...) for a problem of binomial regression (i.e. the
outcome is 0/1, dead or alive etc...).
I have used in the past the multinom function from the nnet library which
uses the neural networks for multinomial regression.
I wonder if I can safely use the same function to investigate a binary
outcome.
I am not providing any code now, as I ask about the applicability of a
library (I mean, meaningful results beside being able to run some code
without any error message) to a slightly less general problem than it has
been designed for.
I would expect the answer to be a solid yes, but... you never know!
Many thanks
Lorenzo
[[alternative HTML version deleted]]
