Kernel smoothers and bootstrapping for semiparametric mixed effects models

JOURNAL OF MULTIVARIATE ANALYSIS

卷:114

Article

Gonzalez Manteiga, Wenceslao²  Jose Lombardia, Maria³  Martinez Miranda, Maria Dolores¹  Sperlich, Stefan^4,5 

[1] Univ Granada, Dept Estadist & IO, E-18071 Granada, Spain

[2] Univ Santiago de Compostela, Dept Estadist & IO, E-15782 Santiago De Compostela, Spain

[3] Univ A Coruna, Dept Matemat, E-15071 La Coruna, Spain

[4] Univ Geneva, Dept Sci Econ, CH-1211 Geneva 4, Switzerland

[5] Univ Geneva, Inst Rech Stat, CH-1211 Geneva 4, Switzerland

关键词: Mixed effects models;    Non- and semiparametric models;    Bootstrap inference;    Bandwidth choice;    Small area statistics;   

DOI  :  10.1016/j.jmva.2012.08.005

来源: Elsevier

【 摘 要 】

While today linear mixed effects models are frequently used tools in different fields of statistics, in particular for studying data with clusters, longitudinal or multi-level structure, the nonparametric formulation of mixed effects models is still quite recent. In this paper we discuss and compare different nonparametric estimation methods. In this context we introduce a computationally inexpensive bootstrap method, which is used to estimate local mean squared errors, to construct confidence intervals and to find locally optimal smoothing parameters. The theoretical considerations are accompanied by the provision of algorithms and simulation studies of the finite sample behavior of the methods. We show that our confidence intervals have good coverage probabilities, and that our bandwidth selection method succeeds to minimize the mean squared error for the nonparametric function locally. (C) 2012 Elsevier Inc. All rights reserved.

【 授权许可】

Free