Search code examples
rregressiongam

Error in as.data.frame.default when fitting a GAM

I am using gam R package. When I use step.Gam() to perform a stepwise regression I get this error:

Error in as.data.frame.default(data, optional = TRUE) : 
  cannot coerce class ‘"function"’ to a data.frame

Here is a reproducible example:

library(dplyr)
library(gam)    

data(mtcars)

df = mtcars %>% mutate(vs = as.factor(vs),
                       am = as.factor(am))

f0 = as.formula("mpg ~ cyl + disp + hp + drat + wt + qsec + vs + am + gear + carb")
m.gam.0 = gam(f0, data = df)
f.gam.step = gam.scope(frame = (df %>% 
                                  select(mpg, cyl, disp, hp, drat, wt, qsec, vs, am, gear, carb) %>%
                                  as.data.frame()),
                       response = 1,
                       arg=c("df=4"))

m.gam.step = step.Gam(object = m.gam.0, scope = f.gam.step, direction="both")

Why do I get this error?

Solution

  • The error is because df is base R function, hence it is treating the variable df as a function rather than dataframe. If you change the dataframe name to something else which is not a function name instead of df it should work.