Fréchet Means in Infinite Dimensions

Adam Quinn Jaffe

Published 2024-10-22Version 1

While there exists a well-developed asymptotic theory of Fr\'echet means of random variables taking values in a general "finite-dimensional" metric space, there are only a few known results in which the random variables can take values in an "infinite-dimensional" metric space. Presently, show that a natural setting for the study of probabilistic aspects of Fr\'echet means is that of metric spaces which admit a suitably powerful notion of "weak convergence". This allows us to recover, strengthen, and generalize virtually all known asymptotic theory for Fr\'echet means; in particular, we expand the possible geometric settings where such theorems can be applied, we reduce the moment assumptions to the provably minimal possible, and we completely remove assumption about uniqueness of the Fr\'echet mean. We also analyze many examples.