Multiple smoothing parameters selection in additive regression quantiles

Vito M.R. Muggeo; Federico Torretta; Paul H.C. Eilers; Mariangela Sciandra; Massimo Attanasio

doi:10.1177/1471082X20929802

Multiple smoothing parameters selection in additive regression quantiles

Vito M.R. Muggeo^*, Federico Torretta, Paul H.C. Eilers, Mariangela Sciandra, Massimo Attanasio

^*Corresponding author for this work

Epidemiology

Research output: Contribution to journal › Article › Academic › peer-review

22 Citations (Scopus)

Abstract

We propose an iterative algorithm to select the smoothing parameters in additive quantile regression, wherein the functional forms of the covariate effects are unspecified and expressed via B-spline bases with difference penalties on the spline coefficients. The proposed algorithm relies on viewing the penalized coefficients as random effects from the symmetric Laplace distribution, and it turns out to be very efficient and particularly attractive with multiple smooth terms. Through simulations we compare our proposal with some alternative approaches, including the traditional ones based on minimization of the Schwarz Information Criterion. A real-data analysis is presented to illustrate the method in practice.

Original language	English
Pages (from-to)	428-448
Number of pages	21
Journal	Statistical Modelling
Volume	21
Issue number	5
Early online date	17 Jul 2021
DOIs	https://doi.org/10.1177/1471082X20929802
Publication status	Published - 1 Oct 2021

Bibliographical note

Funding Information:
Data illustrated in Section 5 have been kindly provided by Professor M. Bellafiore, ?Dipartimento di Scienze Psicologiche, Pedagogiche e della Formazione? University of Palermo.We would like to thank the Editor Professor Arno?t Kom?rek, the Associate Editor and the referees for their suggestions and carefully reading of the manuscript which lead to a substantial improvement of the article.

Publisher Copyright:
© 2020 Statistical Modeling Society.

Access to Document

10.1177/1471082X20929802Licence: Unspecified

Cite this

@article{2086abbe963f43bc8ce87c0d2f65710f,

title = "Multiple smoothing parameters selection in additive regression quantiles",

abstract = "We propose an iterative algorithm to select the smoothing parameters in additive quantile regression, wherein the functional forms of the covariate effects are unspecified and expressed via B-spline bases with difference penalties on the spline coefficients. The proposed algorithm relies on viewing the penalized coefficients as random effects from the symmetric Laplace distribution, and it turns out to be very efficient and particularly attractive with multiple smooth terms. Through simulations we compare our proposal with some alternative approaches, including the traditional ones based on minimization of the Schwarz Information Criterion. A real-data analysis is presented to illustrate the method in practice.",

author = "Muggeo, {Vito M.R.} and Federico Torretta and Eilers, {Paul H.C.} and Mariangela Sciandra and Massimo Attanasio",

note = "Funding Information: Data illustrated in Section 5 have been kindly provided by Professor M. Bellafiore, ?Dipartimento di Scienze Psicologiche, Pedagogiche e della Formazione? University of Palermo.We would like to thank the Editor Professor Arno?t Kom?rek, the Associate Editor and the referees for their suggestions and carefully reading of the manuscript which lead to a substantial improvement of the article. Publisher Copyright: {\textcopyright} 2020 Statistical Modeling Society.",

year = "2021",

month = oct,

day = "1",

doi = "10.1177/1471082X20929802",

language = "English",

volume = "21",

pages = "428--448",

journal = "Statistical Modelling",

issn = "1471-082X",

publisher = "SAGE Publications Ltd",

number = "5",

}

TY - JOUR

T1 - Multiple smoothing parameters selection in additive regression quantiles

AU - Muggeo, Vito M.R.

AU - Torretta, Federico

AU - Eilers, Paul H.C.

AU - Sciandra, Mariangela

AU - Attanasio, Massimo

N1 - Funding Information: Data illustrated in Section 5 have been kindly provided by Professor M. Bellafiore, ?Dipartimento di Scienze Psicologiche, Pedagogiche e della Formazione? University of Palermo.We would like to thank the Editor Professor Arno?t Kom?rek, the Associate Editor and the referees for their suggestions and carefully reading of the manuscript which lead to a substantial improvement of the article. Publisher Copyright: © 2020 Statistical Modeling Society.

PY - 2021/10/1

Y1 - 2021/10/1

N2 - We propose an iterative algorithm to select the smoothing parameters in additive quantile regression, wherein the functional forms of the covariate effects are unspecified and expressed via B-spline bases with difference penalties on the spline coefficients. The proposed algorithm relies on viewing the penalized coefficients as random effects from the symmetric Laplace distribution, and it turns out to be very efficient and particularly attractive with multiple smooth terms. Through simulations we compare our proposal with some alternative approaches, including the traditional ones based on minimization of the Schwarz Information Criterion. A real-data analysis is presented to illustrate the method in practice.

AB - We propose an iterative algorithm to select the smoothing parameters in additive quantile regression, wherein the functional forms of the covariate effects are unspecified and expressed via B-spline bases with difference penalties on the spline coefficients. The proposed algorithm relies on viewing the penalized coefficients as random effects from the symmetric Laplace distribution, and it turns out to be very efficient and particularly attractive with multiple smooth terms. Through simulations we compare our proposal with some alternative approaches, including the traditional ones based on minimization of the Schwarz Information Criterion. A real-data analysis is presented to illustrate the method in practice.

UR - http://www.scopus.com/inward/record.url?scp=85088025792&partnerID=8YFLogxK

U2 - 10.1177/1471082X20929802

DO - 10.1177/1471082X20929802

M3 - Article

AN - SCOPUS:85088025792

SN - 1471-082X

VL - 21

SP - 428

EP - 448

JO - Statistical Modelling

JF - Statistical Modelling

IS - 5

ER -

Multiple smoothing parameters selection in additive regression quantiles

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this