Charles Explorer logo
🇬🇧

Depth for Vector-Valued Functions

Publication at Faculty of Mathematics and Physics |
2013

Abstract

Data depth is a nonparametric statistical tool applicable to multi-dimensional observations. In the contribution a depth suitable for vector-valued functional (infi nite-dimensional) data of one variable is advanced.

Its connections to the established depth functionals for univariate response are explored. It is shown that a number of theoretical results not considered in the literature before may be obtained by a proper application of this generalization technique.