0

I've read all answers and comments here: Why does logistic regression produce well-calibrated models? but still not clear about the answer.

Can someone please elaborate why the following equation means model is well calibrated?

setting derivative of loss function to zero

The two sides of above equation are sum of products. I don't quite understand why we can take enter image description here off the equation and have

enter image description here

Thank you very much!

kjetil b halvorsen
  • 63,378
  • 26
  • 142
  • 467
user253288
  • 3
  • 2
  • 1
    Welcome to CV. Please edit your question (specifically the title) to make it clear how this question differs from the linked question. Follow-up questions are fine, but now your title is identical to the original question, which attracts close votes. – Frans Rodenburg Jul 11 '19 at 01:28
  • Thank you for the advice. It's a follow up question to the linked one, but I don't know how to tag it to make it sounds like a follow up question... – user253288 Jul 11 '19 at 15:42

1 Answers1

0

The first equation applies to every column $x_j$ of the design matrix. In particular, it applies to the first column $x_0$, which is a vector of all ones. That's why the $x$'s drop out in the second equation.

tddevlin
  • 3,205
  • 1
  • 12
  • 27