High-dimensional adaptive function-on-scalar regression

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Applications of functional data with large numbers of predictors have grown precipitously in recent years, driven, in part, by rapid advances in genotyping technologies. Given the large numbers of genetic mutations encountered in genetic association studies, statistical methods which more fully exploit the underlying structure of the data are imperative for maximizing statistical power. However, there is currently very limited work in functional data with large numbers of predictors. Tools are presented for simultaneous variable selection and parameter estimation in a functional linear model with a functional outcome and a large number of scalar predictors; the technique is called AFSL for Adaptive Function-on-Scalar Lasso. It is demonstrated how techniques from convex analysis over Hilbert spaces can be used to establish a functional version of the oracle property for AFSL over any real separable Hilbert space, even when the number of predictors, I, is exponentially large compared to the sample size, N. AFSL is illustrated via a simulation study and data from the Childhood Asthma Management Program, CAMP, selecting those genetic mutations which are important for lung growth.

Original languageEnglish (US)
Pages (from-to)167-183
Number of pages17
JournalEconometrics and Statistics
Volume1
DOIs
StatePublished - Jan 1 2017

Fingerprint

High-dimensional
Regression
Scalar
Predictors
Functional Data
Mutation
Functional Linear Model
Oracle Property
Genetic Association
Asthma
Statistical Power
Convex Analysis
Lasso
Separable Hilbert Space
Variable Selection
Lung
Statistical method
Parameter Estimation
Sample Size
Hilbert space

All Science Journal Classification (ASJC) codes

  • Statistics, Probability and Uncertainty
  • Economics and Econometrics
  • Statistics and Probability

Cite this

@article{a803fa8ca0b7476584fa1af36bcbd5f7,
title = "High-dimensional adaptive function-on-scalar regression",
abstract = "Applications of functional data with large numbers of predictors have grown precipitously in recent years, driven, in part, by rapid advances in genotyping technologies. Given the large numbers of genetic mutations encountered in genetic association studies, statistical methods which more fully exploit the underlying structure of the data are imperative for maximizing statistical power. However, there is currently very limited work in functional data with large numbers of predictors. Tools are presented for simultaneous variable selection and parameter estimation in a functional linear model with a functional outcome and a large number of scalar predictors; the technique is called AFSL for Adaptive Function-on-Scalar Lasso. It is demonstrated how techniques from convex analysis over Hilbert spaces can be used to establish a functional version of the oracle property for AFSL over any real separable Hilbert space, even when the number of predictors, I, is exponentially large compared to the sample size, N. AFSL is illustrated via a simulation study and data from the Childhood Asthma Management Program, CAMP, selecting those genetic mutations which are important for lung growth.",
author = "Zhaohu Fan and Reimherr, {Matthew Logan}",
year = "2017",
month = "1",
day = "1",
doi = "10.1016/j.ecosta.2016.08.001",
language = "English (US)",
volume = "1",
pages = "167--183",
journal = "Econometrics and Statistics",
issn = "2452-3062",
publisher = "Elsevier BV",

}

High-dimensional adaptive function-on-scalar regression. / Fan, Zhaohu; Reimherr, Matthew Logan.

In: Econometrics and Statistics, Vol. 1, 01.01.2017, p. 167-183.

Research output: Contribution to journalArticle

TY - JOUR

T1 - High-dimensional adaptive function-on-scalar regression

AU - Fan, Zhaohu

AU - Reimherr, Matthew Logan

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Applications of functional data with large numbers of predictors have grown precipitously in recent years, driven, in part, by rapid advances in genotyping technologies. Given the large numbers of genetic mutations encountered in genetic association studies, statistical methods which more fully exploit the underlying structure of the data are imperative for maximizing statistical power. However, there is currently very limited work in functional data with large numbers of predictors. Tools are presented for simultaneous variable selection and parameter estimation in a functional linear model with a functional outcome and a large number of scalar predictors; the technique is called AFSL for Adaptive Function-on-Scalar Lasso. It is demonstrated how techniques from convex analysis over Hilbert spaces can be used to establish a functional version of the oracle property for AFSL over any real separable Hilbert space, even when the number of predictors, I, is exponentially large compared to the sample size, N. AFSL is illustrated via a simulation study and data from the Childhood Asthma Management Program, CAMP, selecting those genetic mutations which are important for lung growth.

AB - Applications of functional data with large numbers of predictors have grown precipitously in recent years, driven, in part, by rapid advances in genotyping technologies. Given the large numbers of genetic mutations encountered in genetic association studies, statistical methods which more fully exploit the underlying structure of the data are imperative for maximizing statistical power. However, there is currently very limited work in functional data with large numbers of predictors. Tools are presented for simultaneous variable selection and parameter estimation in a functional linear model with a functional outcome and a large number of scalar predictors; the technique is called AFSL for Adaptive Function-on-Scalar Lasso. It is demonstrated how techniques from convex analysis over Hilbert spaces can be used to establish a functional version of the oracle property for AFSL over any real separable Hilbert space, even when the number of predictors, I, is exponentially large compared to the sample size, N. AFSL is illustrated via a simulation study and data from the Childhood Asthma Management Program, CAMP, selecting those genetic mutations which are important for lung growth.

UR - http://www.scopus.com/inward/record.url?scp=85032798750&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85032798750&partnerID=8YFLogxK

U2 - 10.1016/j.ecosta.2016.08.001

DO - 10.1016/j.ecosta.2016.08.001

M3 - Article

AN - SCOPUS:85032798750

VL - 1

SP - 167

EP - 183

JO - Econometrics and Statistics

JF - Econometrics and Statistics

SN - 2452-3062

ER -