Flexible integrated functional depths

Stanislav Nagy, Sami Helander, Germain van Bever, Lauri Viitasaari, Pauliina Ilmonen

Research output: Contribution to journalArticleScientificpeer-review

1 Citation (Scopus)
2 Downloads (Pure)

Abstract

This paper develops a new class of functional depths. A generic member of this class is coined Jth order kth moment integrated depth. It is based on the distribution of the cross-sectional halfspace depth of a function in the marginal evaluations (in time) of the random process. Asymptotic properties of the proposed depths are provided: we show that they are uniformly consistent and satisfy an inequality related to the law of the iterated logarithm. Moreover, limiting distributions are derived under mild regularity assumptions. The versatility displayed by the new class of depths makes them particularly amenable for capturing important features of functional distributions. This is illustrated in supervised learning, where we show that the corresponding maximum depth classifiers outperform classical competitors.

Original languageEnglish
Pages (from-to)673-701
Number of pages29
JournalBernoulli
Volume27
Issue number1
DOIs
Publication statusPublished - Feb 2021
MoE publication typeA1 Journal article-refereed

Keywords

  • Asymptotics
  • Data depth
  • Functional data analysis
  • Integrated depths
  • Supervised classification

Fingerprint Dive into the research topics of 'Flexible integrated functional depths'. Together they form a unique fingerprint.

Cite this