Warning!
You are viewing this exercise in a course that you are not a member of. You won't appear on your teacher's scoresheets until you register.
Warning! It seems that you are using Dodona within another webpage, so not everything may work properly. Let your teacher know so that he can solve the problem by adjusting a setting in the learning environment. In the meantime, you can click this link to open Dodona in a new window.
Sign in to test your solution.
# we adapted the function a little for the dodona exercise
# for testing your code please use the predict function defined above
# for submitting your code on dodona, use this function instead
predict.regsubsets <- function(object, newdata, id, ...) {
mat <- model.matrix(medv ~ ., data = newdata)
coefi <- coef(object, id = id)
xvars <- names(coefi)
mat[, xvars] %*% coefi
}
k <- ...
folds <- ...
cv.errors <- ...
for (j in 1:...) {
best.fit <- ...
for (i in 1:...) {
pred <- ...
cv.errors[...] <- ...
}
}
mean.cv.errors <- ...
reg.best <- ...
reg.best.coef <- ...
You can submit as many times as you like. Only your latest submission will be taken into account.
Sign in to test your solution.