References & Citations
Mathematics > Statistics Theory
Title: Kernel Selection in Nonparametric Regression
(Submitted on 13 Jun 2020 (v1), last revised 20 Mar 2021 (this version, v2))
Abstract: In the regression model $Y = b(X) +\sigma(X)\varepsilon$, where $X$ has a density $f$, this paper deals with an oracle inequality for an estimator of $bf$, involving a kernel in the sense of Lerasle et al. (2016), selected via the PCO method. In addition to the bandwidth selection for kernel-based estimators already studied in Lacour, Massart and Rivoirard (2017) and Comte and Marie (2020), the dimension selection for anisotropic projection estimators of $f$ and $bf$ is covered.
Submission history
From: Nicolas Marie [view email][v1] Sat, 13 Jun 2020 16:25:27 GMT (410kb,D)
[v2] Sat, 20 Mar 2021 20:09:38 GMT (18kb)
Link back to: arXiv, form interface, contact.