Identification of the constant components in generalised semivarying coefficient models by cross-validation

W Y Zhang

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

In practice, some coefficients in generalised varying coefficient models may be constant. We pay a price on the variance side of an estimator when constant coefficients are treated as special functions. This prompts the question of how to identify the constant coefficients. This is basically a model selection problem. In this paper, we use cross-validation (CV) as a criterion for model selection to identify the constant coefficients. We investigate the asymptotic properties of the proposed CV-based model selection approach. We report on a simulation study conducted to show how well the proposed method works when sample size is finite. Finally, the proposed method is used to analyse a data set from China about contraceptive use, which leads to some interesting findings.
Original languageEnglish
Pages (from-to)1913-1929
Number of pages17
JournalStatistica Sinica
Volume21
Issue number4
DOIs
Publication statusPublished - Oct 2011

Fingerprint

Cross-validation
Model Selection
Coefficient
Varying Coefficient Model
Special Functions
Model
Asymptotic Properties
China
Sample Size
Simulation Study
Estimator
Coefficients
Model selection

Cite this

Identification of the constant components in generalised semivarying coefficient models by cross-validation. / Zhang, W Y.

In: Statistica Sinica, Vol. 21, No. 4, 10.2011, p. 1913-1929.

Research output: Contribution to journalArticle

@article{b229d9c2b0ff4f02b87d390a989d4afb,
title = "Identification of the constant components in generalised semivarying coefficient models by cross-validation",
abstract = "In practice, some coefficients in generalised varying coefficient models may be constant. We pay a price on the variance side of an estimator when constant coefficients are treated as special functions. This prompts the question of how to identify the constant coefficients. This is basically a model selection problem. In this paper, we use cross-validation (CV) as a criterion for model selection to identify the constant coefficients. We investigate the asymptotic properties of the proposed CV-based model selection approach. We report on a simulation study conducted to show how well the proposed method works when sample size is finite. Finally, the proposed method is used to analyse a data set from China about contraceptive use, which leads to some interesting findings.",
author = "Zhang, {W Y}",
year = "2011",
month = "10",
doi = "10.5705/ss.2009.201",
language = "English",
volume = "21",
pages = "1913--1929",
journal = "Statistica Sinica",
issn = "1017-0405",
publisher = "Institute of Statistical Science",
number = "4",

}

TY - JOUR

T1 - Identification of the constant components in generalised semivarying coefficient models by cross-validation

AU - Zhang, W Y

PY - 2011/10

Y1 - 2011/10

N2 - In practice, some coefficients in generalised varying coefficient models may be constant. We pay a price on the variance side of an estimator when constant coefficients are treated as special functions. This prompts the question of how to identify the constant coefficients. This is basically a model selection problem. In this paper, we use cross-validation (CV) as a criterion for model selection to identify the constant coefficients. We investigate the asymptotic properties of the proposed CV-based model selection approach. We report on a simulation study conducted to show how well the proposed method works when sample size is finite. Finally, the proposed method is used to analyse a data set from China about contraceptive use, which leads to some interesting findings.

AB - In practice, some coefficients in generalised varying coefficient models may be constant. We pay a price on the variance side of an estimator when constant coefficients are treated as special functions. This prompts the question of how to identify the constant coefficients. This is basically a model selection problem. In this paper, we use cross-validation (CV) as a criterion for model selection to identify the constant coefficients. We investigate the asymptotic properties of the proposed CV-based model selection approach. We report on a simulation study conducted to show how well the proposed method works when sample size is finite. Finally, the proposed method is used to analyse a data set from China about contraceptive use, which leads to some interesting findings.

UR - http://www.scopus.com/inward/record.url?scp=84863381693&partnerID=8YFLogxK

UR - http://dx.doi.org/10.5705/ss.2009.201

U2 - 10.5705/ss.2009.201

DO - 10.5705/ss.2009.201

M3 - Article

VL - 21

SP - 1913

EP - 1929

JO - Statistica Sinica

JF - Statistica Sinica

SN - 1017-0405

IS - 4

ER -