EUVIMEDEUVIMED entdecken
Crossref

Inference for dependent error functional data: Covariance function

Sijie Zheng, Kun Huang, Lijian Yang

Electronic Journal of Statistics · 01.01.2025 · 19 (2) · 5216-5248

DOI: 10.1214/25-ejs2458 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronic Journal of Statistics, erschienen am 01.01.2025. Verfasst von Sijie Zheng und 2 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 34 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Statistical inference is developed for covariance function of functional data with dependent errors. A two-step B-spline estimation procedure is proposed. The first step involves estimation of all individual trajectories and the second step involves estimation of the covariance function using estimated trajectories in place of the latent true trajectories. Under mild technical assumptions, the covariance estimator is as efficient as an “oracle” estimator if all trajectories were completely observed. Simultaneous confidence envelope (SCE) is constructed for the covariance function, with coverage probability shown to be asymptotically correct. Intensive simulations are conducted to demonstrate the superior theoretical accuracy and performance of covariance estimator and SCE. The proposed SCE is constructed for an evoked electroencephalogram (EEG) data, which leads to simplified and explicit trigonometric form of the functional data.

Zitieren

APA

Zheng, S., Huang, K., & Yang, L. (2025). Inference for dependent error functional data: Covariance function. Electronic Journal of Statistics, 19(2), 5216-5248. https://doi.org/10.1214/25-ejs2458

Vancouver

Zheng S, Huang K, Yang L. Inference for dependent error functional data: Covariance function. Electronic Journal of Statistics. 2025;19(2):5216-5248. doi:10.1214/25-ejs2458

BibTeX
@article{Zheng2025,
  author = {Zheng, Sijie and Huang, Kun and Yang, Lijian},
  title = {Inference for dependent error functional data: Covariance function},
  journal = {Electronic Journal of Statistics},
  year = {2025},
  volume = {19},
  number = {2},
  pages = {5216--5248},
  doi = {10.1214/25-ejs2458}
}
Kostenlos starten →