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: Uniform convergence rates and automatic variabel selection in nonparametric regression with functional and categorical covariates

Authors: Leonie Selk
Abstract: In Selk and Gertheiss (2022) a nonparametric prediction method for models with multiple functional and categorical covariates is introduced. The dependent variable can be categorical (binary or multi-class) or continuous, thus both classification and regression problems are considered. In the paper at hand the asymptotic properties of this method are developed. A uniform rate of convergence for the regression / classification estimator is given. Further it is shown that, asymptotically, a data-driven least squares cross-validation method can automatically remove irrelevant, noise variables.
Subjects: Statistics Theory (math.ST)
MSC classes: 62G08, 62G20, 62H12
Cite as: arXiv:2209.15079 [math.ST]
  (or arXiv:2209.15079v1 [math.ST] for this version)

Submission history

From: Leonie Selk [view email]
[v1] Thu, 29 Sep 2022 20:09:14 GMT (56kb,D)

Link back to: arXiv, form interface, contact.