If I were the one asking you to predict these things, I would want your best subjective probability... especially if other agents might come to different conclusions.
<< modeling the loss (which is minus the loglikelihood) >>
I'm not aware of this loss function. But surely you can define other loss functions, no?
What is curious for me is that the proper scoring rule is not unique.
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
branchandroot
no subject
Date: 2010-11-19 04:10 pm (UTC)<< modeling the loss (which is minus the loglikelihood) >>
I'm not aware of this loss function. But surely you can define other loss functions, no?
What is curious for me is that the proper scoring rule is not unique.