Skip to content
Prev 4170 / 63424 Next

predict.lm is broken in 1.1.0-patched (2000-August-7) (PR#626)

ripley@stats.ox.ac.uk writes:
...
Sometimes the fix for one bug uncovers another. We have
function (object, ...) 
{
    z <- object$coef
    z[!is.na(z)]
}
function (object, ...) 
object$coefficients

and both of these cannot be right when predict.lm calls coef(object)
and object$coefficients contains NA's. Using object$coefficients
inside predict.lm fixes things, but I'm still not happy with the two
different conventions for coef...