Nonparametric feature screening

Lu Lin; Jing Sun; Lixing Zhu

doi:10.1016/j.csda.2013.05.016

Nonparametric feature screening

Lu Lin^*, Jing Sun, Lixing Zhu

^*Corresponding author for this work

Department of Mathematics

Research output: Contribution to journal › Journal article › peer-review

25 Citations (Scopus)

Abstract

The measure of correlation between response and predictors plays a critical role in feature ranking and screening for nonparametric regression models. In this paper, a nonparametric function-correlative feature screening is introduced. The newly proposed method does not need any assumption on structural relationships between response and predictors, and among predictors. By using local information flows of model variables, the function-correlation between response and predictors is captured successfully. Selection consistency is achieved as well. Simulation studies are carried out to examine the performance of the new method.

Original language	English
Pages (from-to)	162-174
Number of pages	13
Journal	Computational Statistics and Data Analysis
Volume	67
DOIs	https://doi.org/10.1016/j.csda.2013.05.016
Publication status	Published - Nov 2013

Scopus Subject Areas

Statistics and Probability
Computational Mathematics
Computational Theory and Mathematics
Applied Mathematics

User-Defined Keywords

Feature screening
Function-correlation
Marginal utility
Nonparametric model
Ultrahigh-dimensional data

Access to Document

10.1016/j.csda.2013.05.016

Cite this

@article{fd68370f534c4add8df5cd2c52ab50cd,

title = "Nonparametric feature screening",

abstract = "The measure of correlation between response and predictors plays a critical role in feature ranking and screening for nonparametric regression models. In this paper, a nonparametric function-correlative feature screening is introduced. The newly proposed method does not need any assumption on structural relationships between response and predictors, and among predictors. By using local information flows of model variables, the function-correlation between response and predictors is captured successfully. Selection consistency is achieved as well. Simulation studies are carried out to examine the performance of the new method.",

keywords = "Feature screening, Function-correlation, Marginal utility, Nonparametric model, Ultrahigh-dimensional data",

author = "Lu Lin and Jing Sun and Lixing Zhu",

note = "Funding Information: Authors thank the referees for the very constructive and thoughtful comments and suggestions. Lu Lin was supported by NNSF project (11171188 and 11231005) of China, Mathematical Finance-Backward Stochastic Analysis and Computations in Financial Risk Control of China ( 11221061 ), NSF and SRRF projects (ZR2010AZ001 and BS2011SF006) of Shandong Province of China. Lixing Zhu was supported by a grant from the University Grants Council of Hong Kong, Hong Kong, China . ",

year = "2013",

month = nov,

doi = "10.1016/j.csda.2013.05.016",

language = "English",

volume = "67",

pages = "162--174",

journal = "Computational Statistics and Data Analysis",

issn = "0167-9473",

publisher = "Elsevier",

}

TY - JOUR

T1 - Nonparametric feature screening

AU - Lin, Lu

AU - Sun, Jing

AU - Zhu, Lixing

N1 - Funding Information: Authors thank the referees for the very constructive and thoughtful comments and suggestions. Lu Lin was supported by NNSF project (11171188 and 11231005) of China, Mathematical Finance-Backward Stochastic Analysis and Computations in Financial Risk Control of China ( 11221061 ), NSF and SRRF projects (ZR2010AZ001 and BS2011SF006) of Shandong Province of China. Lixing Zhu was supported by a grant from the University Grants Council of Hong Kong, Hong Kong, China .

PY - 2013/11

Y1 - 2013/11

N2 - The measure of correlation between response and predictors plays a critical role in feature ranking and screening for nonparametric regression models. In this paper, a nonparametric function-correlative feature screening is introduced. The newly proposed method does not need any assumption on structural relationships between response and predictors, and among predictors. By using local information flows of model variables, the function-correlation between response and predictors is captured successfully. Selection consistency is achieved as well. Simulation studies are carried out to examine the performance of the new method.

AB - The measure of correlation between response and predictors plays a critical role in feature ranking and screening for nonparametric regression models. In this paper, a nonparametric function-correlative feature screening is introduced. The newly proposed method does not need any assumption on structural relationships between response and predictors, and among predictors. By using local information flows of model variables, the function-correlation between response and predictors is captured successfully. Selection consistency is achieved as well. Simulation studies are carried out to examine the performance of the new method.

KW - Feature screening

KW - Function-correlation

KW - Marginal utility

KW - Nonparametric model

KW - Ultrahigh-dimensional data

UR - http://www.scopus.com/inward/record.url?scp=84879231162&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2013.05.016

DO - 10.1016/j.csda.2013.05.016

M3 - Journal article

AN - SCOPUS:84879231162

SN - 0167-9473

VL - 67

SP - 162

EP - 174

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

ER -

Nonparametric feature screening

Abstract

Scopus Subject Areas

User-Defined Keywords

Access to Document

Other files and links

Fingerprint

Cite this