I'm reading Introduction to Statistical Learning, currently Chapter 2 about the Bias-Variance trade-off.

In all examples the irreducible error is 1, i.e. $Var(\epsilon) = 1$. I read in Why is the variance of the error term always 1 in examples of the bias-variance tradeoff? that this is likely for pedagogical reasons.

My question is, how would you actually go about to find the irreducible error? Is this even possible?

  • 1
    $\begingroup$ You don't, it's a conceptual tool. $\endgroup$ Jun 16 '17 at 15:45
  • $\begingroup$ well, it could be defined as the error of the Bayes classifier. I'm not sure but I think there must be ways to estimate it. $\endgroup$
    – synack
    Aug 27 '20 at 0:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.