RT Journal Article
T1 Bandwidth selection for kernel density estimation with length-biased data
A1 Borrajo García, María Isabel
A1 González Manteiga, Wenceslao
A1 Martínez Miranda, María Dolores
K1 Bootstrap
K1 Rule-of-thumb
K1 Cross-validation
K1 Nonparametric
K1 Weighted data
AB Length-biased data are a particular case of weighted data, which arise in many situations: biomedicine, quality control or epidemiology among others. In this paper we study the theoretical properties of kernel density estimation in the context of length-biased data, proposing two consistent bootstrap methods that we use for bandwidth selection. Apart from the bootstrap bandwidth selectors we suggest a rule-of-thumb. These bandwidth selection proposals are compared with a least-squares cross-validation method. A simulation study is accomplished to understand the behaviour of the procedures in finite samples
PB Taylor & Francis
SN 1048-5252
YR 2017
FD 2017
LK http://hdl.handle.net/10347/20313
UL http://hdl.handle.net/10347/20313
LA eng
NO Borrajo, M. I., González-Manteiga, W., & Martínez-Miranda, M. D. (2017). Bandwidth selection for kernel density estimation with length-biased data. Journal of Nonparametric Statistics, 29(3), 636-668.
NO This is an Accepted Manuscript of an article published by Taylor & Francis in Journal of Nonparametric Statistics on 23 Jun 2017, available online: https://doi.org/10.1080/10485252.2017.1339309.
NO The authors acknowledge the support fromthe SpanishMinistry of Economy and Competitiveness, through grant number MTM2013-41383P, which includes support from the European Regional Development Fund (ERDF). Support from the IAP network StUDyS (P7/06) from Belgian Science Policy, is also acknowledged. M.I. Borrajo has been supported by FPU (FPU2013/00473) from the Spanish Ministry of Education
DS Minerva
RD 22 abr 2026