A Nonparametric Bootstrap Method for Heteroscedastic Functional Data
![Thumbnail](/dspace/bitstream/handle/2183/36330/FernandezCasal_Ruben_2024_Nonparametric_bootstrap_method_heteroscedastic.pdf.jpg?sequence=5&isAllowed=y)
Use este enlace para citar
http://hdl.handle.net/2183/36330Coleccións
- GI-MODES - Artigos [141]
Metadatos
Mostrar o rexistro completo do ítemTítulo
A Nonparametric Bootstrap Method for Heteroscedastic Functional DataData
2024-03Resumo
[Absctract]: The objective is to provide a nonparametric bootstrap method for functional data that consists of independent realizations of a continuous one-dimensional process. The process is assumed to be nonstationary, with a functional mean and a functional variance, and dependent. The resampling method is based on nonparametric estimates of the model components. Numerical studies were conducted to check the performance of the proposed procedure, by approximating the bias and the standard error of two estimators. A practical application of the proposed approach to pollution data has also been included. Specifically, it is employed to make inference about the annual trend of ground-level ozone concentration at Yarner Wood monitoring station in the United Kingdom. Supplementary material to this paper is provided online.
Palabras chave
Functional data analysis
Resampling methods
Local linear estimation
Variogram
Resampling methods
Local linear estimation
Variogram
Descrición
Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature The pre-processed data are supplied with the R package npfda, as the ozone data set. The original data can be downloaded from https://uk-air.defra.gov.uk/data. The code used to apply the proposed methodology to the pollution data and the results generated are included in the supplementary material.
Versión do editor
Dereitos
Atribución 3.0 España
ISSN
1537-2693
1085-7117
1085-7117