Comparing FPCA based on conditional quantile functions and FPCA based on conditional mean function

Risultato della ricerca: Chapter

Abstract

In this work functional principal component analysis (FPCA) based on quantile functions is proposed as an alternative to the classical approach, based on the functional mean. Quantile regression characterizes the conditional distribution of a response variable and, in particular, some features like the tails behavior; smoothing splines have also been usefully applied to quantile regression to allow for a more flexible modelling. This framework finds application in contexts involving multiple high frequency time series, for which the functional data analysis (FDA) approach is a natural choice. Quantile regression is then extended to the estimation of functional quantiles and our proposal explores the performance of the three-mode FPCA as a tool for summarizing information when functional quantiles of different order are simultaneously considered. The methodology is illustrated and compared with the functional mean based FPCA through an application to air pollution data.
Lingua originaleEnglish
Titolo della pubblicazione ospiteNew Statistical Developments in Data Science
Pagine65-76
Numero di pagine12
Stato di pubblicazionePublished - 2019

Serie di pubblicazioni

NomeSPRINGER PROCEEDINGS IN MATHEMATICS & STATISTICS

Fingerprint

Functional Principal Component Analysis
Conditional Quantiles
Quantile Function
Quantile Regression
Quantile
Functional Data Analysis
Tail Behavior
Smoothing Splines
Air Pollution
Conditional Distribution
Time series
Methodology
Alternatives
Modeling

All Science Journal Classification (ASJC) codes

  • Mathematics(all)

Cita questo

Di Salvo, F., Ruggieri, M., & Plaia, A. (2019). Comparing FPCA based on conditional quantile functions and FPCA based on conditional mean function. In New Statistical Developments in Data Science (pagg. 65-76). (SPRINGER PROCEEDINGS IN MATHEMATICS & STATISTICS).

Comparing FPCA based on conditional quantile functions and FPCA based on conditional mean function. / Di Salvo, Francesca; Ruggieri, Mariantonietta; Plaia, Antonella.

New Statistical Developments in Data Science. 2019. pag. 65-76 (SPRINGER PROCEEDINGS IN MATHEMATICS & STATISTICS).

Risultato della ricerca: Chapter

Di Salvo, F, Ruggieri, M & Plaia, A 2019, Comparing FPCA based on conditional quantile functions and FPCA based on conditional mean function. in New Statistical Developments in Data Science. SPRINGER PROCEEDINGS IN MATHEMATICS & STATISTICS, pagg. 65-76.
Di Salvo F, Ruggieri M, Plaia A. Comparing FPCA based on conditional quantile functions and FPCA based on conditional mean function. In New Statistical Developments in Data Science. 2019. pag. 65-76. (SPRINGER PROCEEDINGS IN MATHEMATICS & STATISTICS).
Di Salvo, Francesca ; Ruggieri, Mariantonietta ; Plaia, Antonella. / Comparing FPCA based on conditional quantile functions and FPCA based on conditional mean function. New Statistical Developments in Data Science. 2019. pagg. 65-76 (SPRINGER PROCEEDINGS IN MATHEMATICS & STATISTICS).
@inbook{a77fec614c1d4e8eaad590d99d690750,
title = "Comparing FPCA based on conditional quantile functions and FPCA based on conditional mean function",
abstract = "In this work functional principal component analysis (FPCA) based on quantile functions is proposed as an alternative to the classical approach, based on the functional mean. Quantile regression characterizes the conditional distribution of a response variable and, in particular, some features like the tails behavior; smoothing splines have also been usefully applied to quantile regression to allow for a more flexible modelling. This framework finds application in contexts involving multiple high frequency time series, for which the functional data analysis (FDA) approach is a natural choice. Quantile regression is then extended to the estimation of functional quantiles and our proposal explores the performance of the three-mode FPCA as a tool for summarizing information when functional quantiles of different order are simultaneously considered. The methodology is illustrated and compared with the functional mean based FPCA through an application to air pollution data.",
author = "{Di Salvo}, Francesca and Mariantonietta Ruggieri and Antonella Plaia",
year = "2019",
language = "English",
isbn = "978-3-030-21157-8",
series = "SPRINGER PROCEEDINGS IN MATHEMATICS & STATISTICS",
pages = "65--76",
booktitle = "New Statistical Developments in Data Science",

}

TY - CHAP

T1 - Comparing FPCA based on conditional quantile functions and FPCA based on conditional mean function

AU - Di Salvo, Francesca

AU - Ruggieri, Mariantonietta

AU - Plaia, Antonella

PY - 2019

Y1 - 2019

N2 - In this work functional principal component analysis (FPCA) based on quantile functions is proposed as an alternative to the classical approach, based on the functional mean. Quantile regression characterizes the conditional distribution of a response variable and, in particular, some features like the tails behavior; smoothing splines have also been usefully applied to quantile regression to allow for a more flexible modelling. This framework finds application in contexts involving multiple high frequency time series, for which the functional data analysis (FDA) approach is a natural choice. Quantile regression is then extended to the estimation of functional quantiles and our proposal explores the performance of the three-mode FPCA as a tool for summarizing information when functional quantiles of different order are simultaneously considered. The methodology is illustrated and compared with the functional mean based FPCA through an application to air pollution data.

AB - In this work functional principal component analysis (FPCA) based on quantile functions is proposed as an alternative to the classical approach, based on the functional mean. Quantile regression characterizes the conditional distribution of a response variable and, in particular, some features like the tails behavior; smoothing splines have also been usefully applied to quantile regression to allow for a more flexible modelling. This framework finds application in contexts involving multiple high frequency time series, for which the functional data analysis (FDA) approach is a natural choice. Quantile regression is then extended to the estimation of functional quantiles and our proposal explores the performance of the three-mode FPCA as a tool for summarizing information when functional quantiles of different order are simultaneously considered. The methodology is illustrated and compared with the functional mean based FPCA through an application to air pollution data.

UR - http://hdl.handle.net/10447/354896

M3 - Chapter

SN - 978-3-030-21157-8

T3 - SPRINGER PROCEEDINGS IN MATHEMATICS & STATISTICS

SP - 65

EP - 76

BT - New Statistical Developments in Data Science

ER -