Hierarchical adaptive regression kernels for regression with functional predictors

Dawn B. Woodard, Ciprian M Crainiceanu, David Ruppert

Research output: Contribution to journalArticle

Abstract

We propose a new method for regression using a parsimonious and scientifically interpretable representation of functional predictors. Our approach is designed for data that exhibit features such as spikes, dips, and plateaus whose frequency, location, size, and shape varies stochastically across subjects. We propose Bayesian inference of the joint functional and exposure models, and give a method for efficient computation. We contrast our approach with existing state-of-the-art methods for regression with functional predictors, and show that our method is more effective and efficient for data that include features occurring at varying locations. We apply our methodology to a large and complex dataset from the Sleep Heart Health Study, to quantify the association between sleep characteristics and health outcomes. Software and technical appendices are provided in the online supplementary materials.

Original languageEnglish (US)
Pages (from-to)777-800
Number of pages24
JournalJournal of Computational and Graphical Statistics
Volume22
Issue number4
DOIs
StatePublished - 2013

Fingerprint

Kernel Regression
Predictors
Regression
Sleep
Health
Bayesian inference
Spike
Quantify
Vary
Software
Kernel regression
Methodology

Keywords

  • Electroencephalogram
  • Functional data analysis
  • Functional linear model
  • Kernel mixture
  • Lévy adaptive regression kernels
  • Nonparametric Bayes

ASJC Scopus subject areas

  • Discrete Mathematics and Combinatorics
  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Hierarchical adaptive regression kernels for regression with functional predictors. / Woodard, Dawn B.; Crainiceanu, Ciprian M; Ruppert, David.

In: Journal of Computational and Graphical Statistics, Vol. 22, No. 4, 2013, p. 777-800.

Research output: Contribution to journalArticle

@article{37009feb829a453280ea460ec513afbd,
title = "Hierarchical adaptive regression kernels for regression with functional predictors",
abstract = "We propose a new method for regression using a parsimonious and scientifically interpretable representation of functional predictors. Our approach is designed for data that exhibit features such as spikes, dips, and plateaus whose frequency, location, size, and shape varies stochastically across subjects. We propose Bayesian inference of the joint functional and exposure models, and give a method for efficient computation. We contrast our approach with existing state-of-the-art methods for regression with functional predictors, and show that our method is more effective and efficient for data that include features occurring at varying locations. We apply our methodology to a large and complex dataset from the Sleep Heart Health Study, to quantify the association between sleep characteristics and health outcomes. Software and technical appendices are provided in the online supplementary materials.",
keywords = "Electroencephalogram, Functional data analysis, Functional linear model, Kernel mixture, L{\'e}vy adaptive regression kernels, Nonparametric Bayes",
author = "Woodard, {Dawn B.} and Crainiceanu, {Ciprian M} and David Ruppert",
year = "2013",
doi = "10.1080/10618600.2012.694765",
language = "English (US)",
volume = "22",
pages = "777--800",
journal = "Journal of Computational and Graphical Statistics",
issn = "1061-8600",
publisher = "American Statistical Association",
number = "4",

}

TY - JOUR

T1 - Hierarchical adaptive regression kernels for regression with functional predictors

AU - Woodard, Dawn B.

AU - Crainiceanu, Ciprian M

AU - Ruppert, David

PY - 2013

Y1 - 2013

N2 - We propose a new method for regression using a parsimonious and scientifically interpretable representation of functional predictors. Our approach is designed for data that exhibit features such as spikes, dips, and plateaus whose frequency, location, size, and shape varies stochastically across subjects. We propose Bayesian inference of the joint functional and exposure models, and give a method for efficient computation. We contrast our approach with existing state-of-the-art methods for regression with functional predictors, and show that our method is more effective and efficient for data that include features occurring at varying locations. We apply our methodology to a large and complex dataset from the Sleep Heart Health Study, to quantify the association between sleep characteristics and health outcomes. Software and technical appendices are provided in the online supplementary materials.

AB - We propose a new method for regression using a parsimonious and scientifically interpretable representation of functional predictors. Our approach is designed for data that exhibit features such as spikes, dips, and plateaus whose frequency, location, size, and shape varies stochastically across subjects. We propose Bayesian inference of the joint functional and exposure models, and give a method for efficient computation. We contrast our approach with existing state-of-the-art methods for regression with functional predictors, and show that our method is more effective and efficient for data that include features occurring at varying locations. We apply our methodology to a large and complex dataset from the Sleep Heart Health Study, to quantify the association between sleep characteristics and health outcomes. Software and technical appendices are provided in the online supplementary materials.

KW - Electroencephalogram

KW - Functional data analysis

KW - Functional linear model

KW - Kernel mixture

KW - Lévy adaptive regression kernels

KW - Nonparametric Bayes

UR - http://www.scopus.com/inward/record.url?scp=84880971274&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84880971274&partnerID=8YFLogxK

U2 - 10.1080/10618600.2012.694765

DO - 10.1080/10618600.2012.694765

M3 - Article

AN - SCOPUS:84880971274

VL - 22

SP - 777

EP - 800

JO - Journal of Computational and Graphical Statistics

JF - Journal of Computational and Graphical Statistics

SN - 1061-8600

IS - 4

ER -