Department of Economics and Business Economics

Additive regression splines with irrelevant categorical and continuous regressors

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Standard

Additive regression splines with irrelevant categorical and continuous regressors. / Ma, S.; Racine, J.S.

In: Statistica Sinica, Vol. 23, No. 2, 01.04.2013, p. 515-541.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Harvard

APA

CBE

MLA

Vancouver

Author

Ma, S. ; Racine, J.S. / Additive regression splines with irrelevant categorical and continuous regressors. In: Statistica Sinica. 2013 ; Vol. 23, No. 2. pp. 515-541.

Bibtex

@article{fb19e3fa209b425aa5e29240b1dfb54d,
title = "Additive regression splines with irrelevant categorical and continuous regressors",
abstract = "We consider the problem of estimating a relationship using semiparametric additive regression splines when there exist both continuous and categorical regressors, some of which are irrelevant but this is not known a priori. We show that choosing the spline degree, number of subintervals, and bandwidths via cross-validation can automatically remove irrelevant regressors, thereby delivering 'automatic dimension reduction' without the need for pre-testing. Theoretical underpinnings are provided, finite-sample performance is studied, and an illustrative application demonstrates the efficacy of the proposed approach in finite-sample settings. An R package implementing the methods is available from the Comprehensive R Archive Network (Racine and Nie (2011)).",
author = "S. Ma and J.S. Racine",
year = "2013",
month = apr,
day = "1",
doi = "10.5705/ss.2011.096",
language = "English",
volume = "23",
pages = "515--541",
journal = "Statistica Sinica",
issn = "1017-0405",
publisher = "Academia Sinica Institute of Statistical Science",
number = "2",

}

RIS

TY - JOUR

T1 - Additive regression splines with irrelevant categorical and continuous regressors

AU - Ma, S.

AU - Racine, J.S.

PY - 2013/4/1

Y1 - 2013/4/1

N2 - We consider the problem of estimating a relationship using semiparametric additive regression splines when there exist both continuous and categorical regressors, some of which are irrelevant but this is not known a priori. We show that choosing the spline degree, number of subintervals, and bandwidths via cross-validation can automatically remove irrelevant regressors, thereby delivering 'automatic dimension reduction' without the need for pre-testing. Theoretical underpinnings are provided, finite-sample performance is studied, and an illustrative application demonstrates the efficacy of the proposed approach in finite-sample settings. An R package implementing the methods is available from the Comprehensive R Archive Network (Racine and Nie (2011)).

AB - We consider the problem of estimating a relationship using semiparametric additive regression splines when there exist both continuous and categorical regressors, some of which are irrelevant but this is not known a priori. We show that choosing the spline degree, number of subintervals, and bandwidths via cross-validation can automatically remove irrelevant regressors, thereby delivering 'automatic dimension reduction' without the need for pre-testing. Theoretical underpinnings are provided, finite-sample performance is studied, and an illustrative application demonstrates the efficacy of the proposed approach in finite-sample settings. An R package implementing the methods is available from the Comprehensive R Archive Network (Racine and Nie (2011)).

UR - http://www.scopus.com/inward/record.url?scp=84884256580&partnerID=8YFLogxK

U2 - 10.5705/ss.2011.096

DO - 10.5705/ss.2011.096

M3 - Journal article

AN - SCOPUS:84884256580

VL - 23

SP - 515

EP - 541

JO - Statistica Sinica

JF - Statistica Sinica

SN - 1017-0405

IS - 2

ER -