In Elements of Statistical Learning, it distinguishes between "in-sample" and "training" error (Frankly, I found the chapter on errors to be very confusing, especially with how they use the different error terminology). Also see the question asked here: What is the difference between in sample error and training error, and intuition of optimism?

But it seems in a lot situations, when people refer to "in-sample" error in practice, they mean the training error, and the 2 are used interchangeably. Is this typically the case?



Your Answer

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

Browse other questions tagged or ask your own question.