Errors with Predict function and data.frame

Question

I'm new to R, so I'm probably just making a rookie mistake from not understanding the underlying structure of R, but I've read all the questions already on this topic, and still can't seem to get my code to work.

I started here: how to solve predict.lm() error: variable 'affinity' was fitted with type "nmatrix.1" but type "numeric" was supplied

Then I read this: https://stat.ethz.ch/pipermail/r-help/2008-August/170848.html

But I still can't seem to make my prediction value a proper data.frame.

Here's the problem: I'm using the mtcars dataset that comes with R. I've made the basic model:

fitcars<-lm(mpg~wt, data=mtcars)

I then try to do this:

predict(fitcars,data.frame(wt=c(3)),interval=("prediction")) 

to get a confidence interval for the prediction of mpg at a certain weight. However, I continue to get the error:

variable 'center.wt' was fitted with type "nmatrix.1" but type "numeric" was supplied.

It doesn't matter if I first do as.vector(mtcars$wt).

I even tried to do this fix, as suggested in the above posts:

fitcars<-lm(mpg~as.vector(center.wt), data=mtcars) 

And I just got this error:

Error in model.frame.default(formula = mpg ~ as.vector(center.wt), data = mtcars,  : 
  invalid type (list) for variable 'as.vector(center.wt)'

I'm stumped, and know I'm just making a stupid mistake. Could anyone help?

Thank you!!

Answer 1

Hm, I seem to have no problem with:

fitcars<-lm(mpg~wt,data=mtcars)
new<-data.frame(wt=2.5)
predict(fitcars,new,interval='prediction')