Partially linear functional quantile regression in a reproducing kernel Hilbert space
Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review
Author(s)
Related Research Unit(s)
Detail(s)
Original language | English |
---|---|
Pages (from-to) | 789–803 |
Number of pages | 15 |
Journal / Publication | Journal of Nonparametric Statistics |
Volume | 34 |
Issue number | 4 |
Online published | 19 May 2022 |
Publication status | Published - Dec 2022 |
Link(s)
Abstract
We consider quantile functional regression with a functional part and a scalar linear part. We establish the optimal prediction rate for the model under mild assumptions in the reproducing kernel Hilbert space (RKHS) framework. Under stronger assumptions related to the capacity of the RKHS, the non-functional linear part is shown to have asymptotic normality. The estimators are illustrated in simulation studies.
Research Area(s)
- Convergence rate, prediction risk, quantile regression, rademacher complexity, MODELS, PREDICTION
Citation Format(s)
Partially linear functional quantile regression in a reproducing kernel Hilbert space. / Zhou, Yan; Zhang, Weiping; Lin, Hongmei et al.
In: Journal of Nonparametric Statistics, Vol. 34, No. 4, 12.2022, p. 789–803.
In: Journal of Nonparametric Statistics, Vol. 34, No. 4, 12.2022, p. 789–803.
Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review