CODEWITHSUNDEEP
r
user2737555
user2737555

Reputation:

Warning message in predict.gam - why is this happening?

Reference: Page 22 of, http://www.statoek.wiso.uni-goettingen.de/mitarbeiter/ogi/pub/r_workshop.pdf

My question is self-contained in this post though. The data is in http://134.76.173.220/R_workshop

car <- read.table("car.dat", header = TRUE)
attach(car)

library(mgcv)
fit <- gam(MPG~s(SP))


plot(HP, MPG)
x <- seq(0, 350, length = 500)
y <- predict(fit, data.frame(HP = x))
lines(x, y, col = "red", lwd = 2)

Error im receiving:

Warning message:
In predict.gam(fit, data.frame(HP = x)) :
  not all required variables have been supplied in  newdata!

Upvotes: 3

Views: 2726

Answers (2)

Stephen Henderson
Stephen Henderson

Reputation: 6532

I think it is supposed to be MPG~HP as that is a good fit and SP is poor (with default parameters)

par(mfrow=c(1,2))

fit <- gam(MPG~s(SP), data=car)
plot(car$SP, car$MPG)
x <- seq(0, 350, length = 500)
y <- predict(fit, data.frame(SP = x))
lines(x, y, col = "red", lwd = 2)

fit <- gam(MPG~s(HP), data=car)
plot(car$HP, car$MPG)
x <- seq(0, 350, length = 500)
y <- predict(fit, data.frame(HP = x))
lines(x, y, col = "red", lwd = 2)

bad fit and good fit

Upvotes: 0

Ed G
Ed G

Reputation: 822

It seems to work if you scrap the 'attach' adjust the code accordingly and use SP as suggested by @GeorgeDontas

fit <- gam(MPG~s(SP), data=car)

plot(car$HP, car$MPG)
x <- seq(0, 350, length = 500)
y <- predict(fit, data.frame(SP = x))
lines(x, y, col = "red", lwd = 2)

Upvotes: 3

Related Questions