A Note on the Gaussian Minimum Conjecture

Abstract: Let

n g e q 2

and

(X_{i}, 1 l e q i l e q n)

be a centered Gaussian random vector. The Gaussian minimum conjecture says that

E l e f t (m i n_{1 l e q i l e q n} | X_{i} | i g h t) g e q E l e f t (m i n_{1 l e q i l e q n} | Y_{i} | i g h t)

, where

Y_{1}, l d o t s, Y_{n}

are independent centered Gaussian random variables with

E (X_{i}^{2}) = E (Y_{i}^{2})

for any

i = 1, l d o t s, n

. In this note, we will show that this conjecture holds if and only if

n = 2

.

This page was built for publication: A Note on the Gaussian Minimum Conjecture