Information-theoretic convergence of extreme values to the Gumbel distribution

07/07/2020
by   Oliver Johnson, et al.
0

We show how convergence to the Gumbel distribution in an extreme value setting can be understood in an information-theoretic sense. We introduce a new type of score function which behaves well under the maximum operation, and which implies simple expressions for entropy and relative entropy. We show that, assuming certain properties of the von Mises representation, convergence to the Gumbel can be proved in the strong sense of relative entropy.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset