Distances and inference for covariance operators

A framework is developed for inference concerning the covariance operator of a functional random process, where the covariance operator itself is an object of interest for statistical analysis. Distances for comparing positive-definite covariance matrices are either extended or shown to be inapplica...

Full description

Bibliographic Details
Main Authors: Pigoli, Davide, Aston, John A.D., Dryden, Ian L., Secchi, Piercesare
Format: Article
Published: Oxford University Press 2014
Subjects:
Online Access:https://eprints.nottingham.ac.uk/41017/
_version_ 1848796179194707968
author Pigoli, Davide
Aston, John A.D.
Dryden, Ian L.
Secchi, Piercesare
author_facet Pigoli, Davide
Aston, John A.D.
Dryden, Ian L.
Secchi, Piercesare
author_sort Pigoli, Davide
building Nottingham Research Data Repository
collection Online Access
description A framework is developed for inference concerning the covariance operator of a functional random process, where the covariance operator itself is an object of interest for statistical analysis. Distances for comparing positive-definite covariance matrices are either extended or shown to be inapplicable to functional data. In particular, an infinite-dimensional analogue of the Procrustes size-and-shape distance is developed. Convergence of finite-dimensional approximations to the infinite-dimensional distance metrics is also shown. For inference, a Fréchet estimator of both the covariance operator itself and the average covariance operator is introduced. A permutation procedure to test the equality of the covariance operators between two groups is also considered. Additionally, the use of such distances for extrapolation to make predictions is explored. As an example of the proposed methodology, the use of covariance operators has been suggested in a philological study of cross-linguistic dependence as a way to incorporate quantitative phonetic information. It is shown that distances between languages derived from phonetic covariance functions can provide insight into the relationships between the Romance languages.
first_indexed 2025-11-14T19:43:52Z
format Article
id nottingham-41017
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:43:52Z
publishDate 2014
publisher Oxford University Press
recordtype eprints
repository_type Digital Repository
spelling nottingham-410172020-05-04T16:46:34Z https://eprints.nottingham.ac.uk/41017/ Distances and inference for covariance operators Pigoli, Davide Aston, John A.D. Dryden, Ian L. Secchi, Piercesare A framework is developed for inference concerning the covariance operator of a functional random process, where the covariance operator itself is an object of interest for statistical analysis. Distances for comparing positive-definite covariance matrices are either extended or shown to be inapplicable to functional data. In particular, an infinite-dimensional analogue of the Procrustes size-and-shape distance is developed. Convergence of finite-dimensional approximations to the infinite-dimensional distance metrics is also shown. For inference, a Fréchet estimator of both the covariance operator itself and the average covariance operator is introduced. A permutation procedure to test the equality of the covariance operators between two groups is also considered. Additionally, the use of such distances for extrapolation to make predictions is explored. As an example of the proposed methodology, the use of covariance operators has been suggested in a philological study of cross-linguistic dependence as a way to incorporate quantitative phonetic information. It is shown that distances between languages derived from phonetic covariance functions can provide insight into the relationships between the Romance languages. Oxford University Press 2014-04-17 Article PeerReviewed Pigoli, Davide, Aston, John A.D., Dryden, Ian L. and Secchi, Piercesare (2014) Distances and inference for covariance operators. Biometrika, 101 (2). pp. 409-422. ISSN 1464-3510 Distance metric; Functional data analysis; Procrustes analysis; Shape analysis https://academic.oup.com/biomet/article-lookup/doi/10.1093/biomet/asu008 doi:10.1093/biomet/asu008 doi:10.1093/biomet/asu008
spellingShingle Distance metric; Functional data analysis; Procrustes analysis; Shape analysis
Pigoli, Davide
Aston, John A.D.
Dryden, Ian L.
Secchi, Piercesare
Distances and inference for covariance operators
title Distances and inference for covariance operators
title_full Distances and inference for covariance operators
title_fullStr Distances and inference for covariance operators
title_full_unstemmed Distances and inference for covariance operators
title_short Distances and inference for covariance operators
title_sort distances and inference for covariance operators
topic Distance metric; Functional data analysis; Procrustes analysis; Shape analysis
url https://eprints.nottingham.ac.uk/41017/
https://eprints.nottingham.ac.uk/41017/
https://eprints.nottingham.ac.uk/41017/