Variable selection in heteroscedastic single-index quantile regression

Eliana Christou, Michael G. Akritas

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

We propose a new algorithm for simultaneous variable selection and parameter estimation for the single-index quantile regression (SIQR) model. The proposed algorithm, which is non iterative, consists of two steps. Step 1 performs an initial variable selection method. Step 2 uses the results of Step 1 to obtain better estimation of the conditional quantiles and, using them, to perform simultaneous variable selection and estimation of the parametric component of the SIQR model. It is shown that the initial variable selection method consistently estimates the relevant variables, and the estimated parametric component derived in Step 2 satisfies the oracle property.

Original languageEnglish (US)
Pages (from-to)6019-6033
Number of pages15
JournalCommunications in Statistics - Theory and Methods
Volume47
Issue number24
DOIs
StatePublished - Dec 17 2018

Fingerprint

Quantile Regression
Variable Selection
Regression Model
Oracle Property
Conditional Quantiles
Parameter Estimation
Estimate

All Science Journal Classification (ASJC) codes

  • Statistics and Probability

Cite this

@article{b0b0e47efeec4f579912a006b5235f62,
title = "Variable selection in heteroscedastic single-index quantile regression",
abstract = "We propose a new algorithm for simultaneous variable selection and parameter estimation for the single-index quantile regression (SIQR) model. The proposed algorithm, which is non iterative, consists of two steps. Step 1 performs an initial variable selection method. Step 2 uses the results of Step 1 to obtain better estimation of the conditional quantiles and, using them, to perform simultaneous variable selection and estimation of the parametric component of the SIQR model. It is shown that the initial variable selection method consistently estimates the relevant variables, and the estimated parametric component derived in Step 2 satisfies the oracle property.",
author = "Eliana Christou and Akritas, {Michael G.}",
year = "2018",
month = "12",
day = "17",
doi = "10.1080/03610926.2017.1405271",
language = "English (US)",
volume = "47",
pages = "6019--6033",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",
number = "24",

}

Variable selection in heteroscedastic single-index quantile regression. / Christou, Eliana; Akritas, Michael G.

In: Communications in Statistics - Theory and Methods, Vol. 47, No. 24, 17.12.2018, p. 6019-6033.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Variable selection in heteroscedastic single-index quantile regression

AU - Christou, Eliana

AU - Akritas, Michael G.

PY - 2018/12/17

Y1 - 2018/12/17

N2 - We propose a new algorithm for simultaneous variable selection and parameter estimation for the single-index quantile regression (SIQR) model. The proposed algorithm, which is non iterative, consists of two steps. Step 1 performs an initial variable selection method. Step 2 uses the results of Step 1 to obtain better estimation of the conditional quantiles and, using them, to perform simultaneous variable selection and estimation of the parametric component of the SIQR model. It is shown that the initial variable selection method consistently estimates the relevant variables, and the estimated parametric component derived in Step 2 satisfies the oracle property.

AB - We propose a new algorithm for simultaneous variable selection and parameter estimation for the single-index quantile regression (SIQR) model. The proposed algorithm, which is non iterative, consists of two steps. Step 1 performs an initial variable selection method. Step 2 uses the results of Step 1 to obtain better estimation of the conditional quantiles and, using them, to perform simultaneous variable selection and estimation of the parametric component of the SIQR model. It is shown that the initial variable selection method consistently estimates the relevant variables, and the estimated parametric component derived in Step 2 satisfies the oracle property.

UR - http://www.scopus.com/inward/record.url?scp=85036639888&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85036639888&partnerID=8YFLogxK

U2 - 10.1080/03610926.2017.1405271

DO - 10.1080/03610926.2017.1405271

M3 - Article

AN - SCOPUS:85036639888

VL - 47

SP - 6019

EP - 6033

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 24

ER -