Translator Disclaimer
February 2019 Functional data analysis by matrix completion
Marie-Hélène Descary, Victor M. Panaretos
Ann. Statist. 47(1): 1-38 (February 2019). DOI: 10.1214/17-AOS1590


Functional data analyses typically proceed by smoothing, followed by functional PCA. This paradigm implicitly assumes that rough variation is due to nuisance noise. Nevertheless, relevant functional features such as time-localised or short scale fluctuations may indeed be rough relative to the global scale, but still smooth at shorter scales. These may be confounded with the global smooth components of variation by the smoothing and PCA, potentially distorting the parsimony and interpretability of the analysis. The goal of this paper is to investigate how both smooth and rough variations can be recovered on the basis of discretely observed functional data. Assuming that a functional datum arises as the sum of two uncorrelated components, one smooth and one rough, we develop identifiability conditions for the recovery of the two corresponding covariance operators. The key insight is that they should possess complementary forms of parsimony: one smooth and finite rank (large scale), and the other banded and potentially infinite rank (small scale). Our conditions elucidate the precise interplay between rank, bandwidth and grid resolution. Under these conditions, we show that the recovery problem is equivalent to rank-constrained matrix completion, and exploit this to construct estimators of the two covariances, without assuming knowledge of the true bandwidth or rank; we study their asymptotic behaviour, and then use them to recover the smooth and rough components of each functional datum by best linear prediction. As a result, we effectively produce separate functional PCAs for smooth and rough variation.


Download Citation

Marie-Hélène Descary. Victor M. Panaretos. "Functional data analysis by matrix completion." Ann. Statist. 47 (1) 1 - 38, February 2019.


Received: 1 August 2016; Revised: 1 February 2017; Published: February 2019
First available in Project Euclid: 30 November 2018

zbMATH: 07036193
MathSciNet: MR3909925
Digital Object Identifier: 10.1214/17-AOS1590

Primary: 62M
Secondary: 15A99, 60G17, 62M15

Rights: Copyright © 2019 Institute of Mathematical Statistics


This article is only available to subscribers.
It is not available for individual sale.

Vol.47 • No. 1 • February 2019
Back to Top