My model (Logistic Regression) has AUC score of 0.8

Am I right in stating that the probability of the model ranking a random positive sample higher than a random negative sample is 0.8? Also, how does this interpretation change if I pick more than 2 random samples (let's say 5)? Can I say that the probability of the model to predict highest probability for the best sample (positive) out of 5 is 0.8? Or will this become 0.8*0.8*0.8*0.8 = 0.4096?

  • $\begingroup$ what do you mean by "best sample"? $\endgroup$ – oW_ Jul 23 '19 at 22:30
  • $\begingroup$ I meant the one which actually has the highest probability of success (1) out of the 5 samples $\endgroup$ – skppit19 Jul 23 '19 at 22:33

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.