Goldburd_ConfMatrix.pdf
The pdf says that the number of at a threshold of 0.55 is less than the number of false positives at 0.30. Unless I'm misreading, the number of false positives doesn't decrease until the threshold hits 80%.
The pdf says that the number of at a threshold of 0.55 is less than the number of false positives at 0.30. Unless I'm misreading, the number of false positives doesn't decrease until the threshold hits 80%.
Comments
Thanks for pointing this out. You're correct.
Change claim number 4 to be N with 52% predicted probability of going to litigation and you get what we were aiming to show. We're updating the files now.