A partially linear single-index transformation model and its nonparametric estimation
被引:5
|
作者:
Ding, Xiaobo
论文数: 0引用数: 0
h-index: 0
机构:
Chinese Acad Sci, Acad Math & Syst Sci, Beijing, Peoples R China
Vet Affairs Puget Sound Hlth Care Syst, HSR& Ctr Excellence, Seattle, WA 98108 USA
Univ Washington, Dept Biostat, Seattle, WA 98195 USAChinese Acad Sci, Acad Math & Syst Sci, Beijing, Peoples R China
Ding, Xiaobo
[1
,2
,3
]
Zhou, Xiao-Hua
论文数: 0引用数: 0
h-index: 0
机构:
Vet Affairs Puget Sound Hlth Care Syst, HSR& Ctr Excellence, Seattle, WA 98108 USA
Univ Washington, Dept Biostat, Seattle, WA 98195 USAChinese Acad Sci, Acad Math & Syst Sci, Beijing, Peoples R China
Zhou, Xiao-Hua
[2
,3
]
Wang, Qihua
论文数: 0引用数: 0
h-index: 0
机构:
Chinese Acad Sci, Acad Math & Syst Sci, Beijing, Peoples R China
Shenzhen Univ, Inst Stat Sci, Shenzhen, Guangdong, Peoples R ChinaChinese Acad Sci, Acad Math & Syst Sci, Beijing, Peoples R China
Wang, Qihua
[1
,4
]
机构:
[1] Chinese Acad Sci, Acad Math & Syst Sci, Beijing, Peoples R China
[2] Vet Affairs Puget Sound Hlth Care Syst, HSR& Ctr Excellence, Seattle, WA 98108 USA
[3] Univ Washington, Dept Biostat, Seattle, WA 98195 USA
[4] Shenzhen Univ, Inst Stat Sci, Shenzhen, Guangdong, Peoples R China
Basis function;
Cross validation;
linear inequality restriction;
local linear regression;
minimum average variance estimation;
SEMIPARAMETRIC ESTIMATION;
DIMENSION REDUCTION;
REGRESSION-MODEL;
D O I:
10.1002/cjs.11239
中图分类号:
O21 [概率论与数理统计];
C8 [统计学];
学科分类号:
020208 ;
070103 ;
0714 ;
摘要:
In this paper, we consider the nonparametric estimation of the partially linear single-index transformation model, where the transformation function, single-index function and error distribution are all completely unknown. We first use the minimum average variance estimation method to estimate the regression coefficients, and then propose a new incorporated local linear regression estimator for the derivative function of the single-index function. Accordingly by integration we can obtain the estimator of the single-index function. Finally we propose a constrained least square estimator for the transformation function, where basis function approximation is employed and cross validation method is proposed to select suitable sets of basis functions. Asymptotical properties of the estimators are established. Simulation studies show that our proposed estimators work well. A real-world data analysis of total health care charges was used to illustrate the proposed procedure. The Canadian Journal of Statistics 43: 97-117; 2015 (c) 2015 Statistical Society of Canada Resume Les auteurs traitent de l'estimation non parametrique du modele de transformation partiellement lineaire a un indice dans le cas oU la transformation, la fonction d'indice et la distribution de l'erreur sont toutes completement inconnues. Ils estiment d'abord les coefficients de regression par la methode de la variance moyenne minimale, puis ils proposent un nouvel estimateur local de regression lineaire pour la derivee de la fonction d'indice, menant a l'estimation de cette fonction par integration. Finalement, ils proposent un estimateur aux moindres carres avec contraintes pour la fonction de transformation dont une approximation est effectuee a l'aide de fonctions de base, elles-memes selectionnees par une procedure de validation croisee. Les auteurs etablissent les proprietes asymptotiques de leurs estimateurs et montrent a l'aide de simulations que ceux-ci donnent de bons resultats. Ils illustrent egalement leur methode en analysant un jeu de donnees reelles portant sur les couts des soins de sante. La revue canadienne de statistique 43: 97-117; 2015 (c) 2015 Societe statistique du Canada
机构:
Univ Strasbourg, IRMA, UMR 7501, 7 Rue Rene Descartes, F-67084 Strasbourg, FranceUniv Strasbourg, IRMA, UMR 7501, 7 Rue Rene Descartes, F-67084 Strasbourg, France
Maistre, Samuel
Patilea, Valentin
论文数: 0引用数: 0
h-index: 0
机构:
ENSAI, Campus Ker Lann,Rue Blaise Pascal,BP 37203, F-35712 Bruz, FranceUniv Strasbourg, IRMA, UMR 7501, 7 Rue Rene Descartes, F-67084 Strasbourg, France
机构:
Hong Kong Baptist Coll, Dept Math, Stat Res & Consultancy Ct, Hong Kong, Hong Kong, Peoples R ChinaHong Kong Baptist Coll, Dept Math, Stat Res & Consultancy Ct, Hong Kong, Hong Kong, Peoples R China
Zhu, LX
Xue, LG
论文数: 0引用数: 0
h-index: 0
机构:Hong Kong Baptist Coll, Dept Math, Stat Res & Consultancy Ct, Hong Kong, Hong Kong, Peoples R China
机构:
Univ York, Dept Econ & Related Studies, York YO10 5DD, N Yorkshire, England
Univ Queensland, Sch Math & Phys, Brisbane, Qld 4072, AustraliaUniv York, Dept Econ & Related Studies, York YO10 5DD, N Yorkshire, England
Chen, Jia
论文数: 引用数:
h-index:
机构:
Gao, Jiti
Li, Degui
论文数: 0引用数: 0
h-index: 0
机构:
Monash Univ, Dept Econometr & Business Stat, Caulfield, Vic 3145, Australia
Univ York, Dept Math, York YO10 5DD, N Yorkshire, EnglandUniv York, Dept Econ & Related Studies, York YO10 5DD, N Yorkshire, England
机构:
E China Normal Univ, Dept Stat & Actuarial Sci, Shanghai 200241, Peoples R ChinaE China Normal Univ, Dept Stat & Actuarial Sci, Shanghai 200241, Peoples R China
Huang, Zhensheng
Zhang, Riquan
论文数: 0引用数: 0
h-index: 0
机构:
Shanxi Datong Univ, Dept Math, Datong 037009, Peoples R ChinaE China Normal Univ, Dept Stat & Actuarial Sci, Shanghai 200241, Peoples R China
Zhang, Riquan
INDIAN JOURNAL OF PURE & APPLIED MATHEMATICS,
2009,
40
(02):
: 87
-
111