I'm currently working on a logistic regression analysis and want to determine if my model validates well. I used the following R code using the "boot" package:

results <- glm(R0A1 ~ SeasonNew * (MP_Scaled + MPHW_Scaled + HW_Scaled + YP_Scaled + AG_Scaled + Shrub_Scaled), family = binomial, data = turkey2nd)
cost <- function(r, pi = 0) mean(abs(r - pi ) > 0.5)
(output <- cv.glm(turkey2nd, results, K=10)$delta)

My estimated delta was 0.23; therefore, is the correct classification 1 - 0.23 = 0.77 or 77% accuracy? Thanks for the assistance!


1 Answer 1



The output of the cv.glm function in the case of binary classification is the estimation of the classification error rate on the test set (it is computed by averaging the classification error rate of the corresponding K validation sets).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.