Limit theorems for Fréchet mean sets

Abstract: For

1 l e p < i n f t y

, the Fr'echet

p

-mean of a probability distribution

m u

on a metric space

(X, d)

is the set

F_{p} (m u) := {a r g, m i n}_{x i n X} i n t_{X} d^{p} (x, y), d m u (y)

, which is taken to be empty if no minimizer exists. Given a sequence

(Y_{i})_{i i n m a t h b b N}

of independent, identically distributed random samples from some probability measure

m u

on

X

, the Fr'echet

p

-means of the empirical measures,

F_{p} (f r a c 1 n s u m_{i = 1}^{n} d e l t a_{Y_{i}})

form a sequence of random closed subsets of

X

. We investigate the senses in which this sequence of random closed sets and related objects converge almost surely as

n o i n f t y

.

Cites work

Cited in

(11)

This page was built for publication: Limit theorems for Fréchet mean sets