EUVIMEDEUVIMED entdecken
Open Access

Functional principal component analysis forsparse censored data

Caitrin Murphy, Eric Laber iD, Rhonda Merwin, Brian Reich iD, Jake Koerner

Biometrika · 13.02.2026 · 113 (2)

DOI: 10.1093/biomet/asag023 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biometrika, online erschienen am 16.04.2026. Verfasst von Caitrin Murphy und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistical Science, Duke University, 214 Old Chemistry, Durham, North Carolina 27708, U.S.A. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 32 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NCI NIH HHS, NSF DMS, NIH.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Functional principal component analysis is a key tool in the study of functional data, driving both exploratory analyses and feature construction for use in formal modelling and testing procedures. However, existing methods do not apply when functional observations are censored; for example, when the measurement instrument only supports recordings within a prespecified interval, thereby truncating values outside this range to the nearest boundary. A naïve application of existing methods, without correction for instrument-induced censoring, introduces bias into the estimators of the mean, covariance and functional principal component scores. We extend the functional principal component analysis framework to accommodate noisy and potentially sparse censored functional data. Local loglikelihood maximization is used to recover smooth estimates of the mean and covariance surface that are representative of the latent process’s mean and covariance functions. The covariance-smoothing procedure yields a positive semidefinite covariance surface, computed without the need to retroactively remove negative eigenvalues in the covariance-operator decomposition. Additionally, we construct a predictor of the scores, conditional on the censored functional data, and demonstrate its use in the generalized functional linear model. Convergence rates for the proposed estimators are established. In simulation experiments, the proposed method yields improved predictive performance and lower bias than existing alternatives. We illustrate its practical value in a study aimed at classifying eating disorder diagnoses in individuals with type 1 diabetes, using censored functional blood glucose data.

Schlagwörter

Stichwörter der Autor:innen

Censored functional dataCensored predictorFunctional principal component analysisScalar-on-function regression

Autor:innen und Institutionen

  • Department of Statistical Science, Duke University, 214 Old Chemistry, Durham, North Carolina 27708, U.S.A.
  • Department of Psychiatry & Behavioral Sciences, Duke University, 905 W Main St, Durham, North Carolina 27701, U.S.A.
  • Department of Statistics, North Carolina State University, 2311 Stinson Drive, Raleigh, North Carolina 27695, U.S.A.

ORCID-Profile: Eric Laber, Brian Reich

Förderung

  • NCI NIH HHS
  • NSF DMS
  • NIH

Zitieren

APA

Murphy, C., Laber, E., Merwin, R., Reich, B., & Koerner, J. (2026). Functional principal component analysis forsparse censored data. Biometrika, 113(2). https://doi.org/10.1093/biomet/asag023

Vancouver

Murphy C, Laber E, Merwin R, Reich B, Koerner J. Functional principal component analysis forsparse censored data. Biometrika. 2026;113(2). doi:10.1093/biomet/asag023

BibTeX
@article{Murphy2026,
  author = {Murphy, Caitrin and Laber, Eric and Merwin, Rhonda and Reich, Brian and Koerner, Jake},
  title = {Functional principal component analysis forsparse censored data},
  journal = {Biometrika},
  year = {2026},
  volume = {113},
  number = {2},
  doi = {10.1093/biomet/asag023}
}
Kostenlos starten →