Box Tidwell - errors coming up

Asked Mar 02 '14 at 01:04

Active Mar 02 '14 at 11:36

Viewed 1,142 times

I'm using:

library(car)
boxTidwell(y, ...)

mydata = my data frame

When I try:

boxTidwell(response~variable1+variable2+...varibalen, ~categoricalvariable, data=mydata)

it comes up with this error:

Error in lm.fit(cbind(1, x1.p, x2), y, ...) : NA/NaN/Inf in 'x'

I run boxTidwell(y,x3,x2=NULL) and get the following error:

Error in model.frame.default(formula = y ~ cbind(x1, x2), drop.unused.levels = TRUE) : 
  invalid type (NULL) for variable 'y'

where:

y = mydata$response 
x1 = mydata[,-response] then x3 = x1[,-categoricalvariable] (i used head(x1) and head(x3) to check this worked)
x2 = mydata$categoricalvariable (i only have one so this can be done)

the categorical variable or the response is not included in x1.

I have read the help file but am really not sure whats wrong. I also tried converting the above to matrix form (model.matrix) instead of data frames (i checked using class) but that threw up the same errors as above.

Any ideas on why this isn't working and how I can fix it?

edited Mar 02 '14 at 11:36

Thomas

43,637
12
109
140

asked Mar 02 '14 at 01:04

Dino Abraham

can you post a reproducible example - perhaps with a snippet of your data? – Peter Ellis Mar 02 '14 at 05:14
Per @PeterEllis, read about making [a reproducible example](http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example). – Thomas Mar 02 '14 at 11:37
How do I attach the CSV file please? – Dino Abraham Mar 02 '14 at 12:56
I have made an example but how do I upload it here please? – Dino Abraham Mar 03 '14 at 14:13

Box Tidwell - errors coming up

0 Answers0