We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

math.ST

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Mathematics > Statistics Theory

Title: Nonparametric regression based on discretely sampled curves

Abstract: In the context of nonparametric regression, we study conditions under which the consistency (and rates of convergence) of estimators built from discretely sampled curves can be derived from the consistency of estimators based on the unobserved whole trajectories. As a consequence, we derive asymptotic results for most of the regularization techniques used in functional data analysis, including smoothing and basis representation.
Subjects: Statistics Theory (math.ST)
Cite as: arXiv:1604.08428 [math.ST]
  (or arXiv:1604.08428v3 [math.ST] for this version)

Submission history

From: Pamela Llop [view email]
[v1] Thu, 28 Apr 2016 14:15:27 GMT (20kb)
[v2] Thu, 6 Oct 2016 17:06:02 GMT (18kb)
[v3] Fri, 26 May 2017 15:21:44 GMT (18kb)

Link back to: arXiv, form interface, contact.