Estimation for High-Dimensional Linear Mixed-Effects Models Using l1-Penalization

被引：115

作者：

Schelldorfer, Juerg ^{[1
]}

Buehlmann, Peter ^{[1
]}

Van De Geer, Sara ^{[1
]}

机构：

[1] ETH, Dept Math, CH-8092 Zurich, Switzerland

来源：

SCANDINAVIAN JOURNAL OF STATISTICS | 2011年 / 38卷 / 02期

基金：

瑞士国家科学基金会;

关键词：

adaptive Lasso; coordinate gradient descent; coordinatewise optimization; Lasso; random-effects model; variable selection; variance components; VARIABLE SELECTION; COORDINATE DESCENT; REGRESSION-MODELS; DANTZIG SELECTOR; ADAPTIVE LASSO; SPARSITY;

D O I：

10.1111/j.1467-9469.2011.00740.x

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：

We propose an l(1)-penalized estimation procedure for high-dimensional linear mixed-effects models. The models are useful whenever there is a grouping structure among high-dimensional observations, that is, for clustered data. We prove a consistency and an oracle optimality result and we develop an algorithm with provable numerical convergence. Furthermore, we demonstrate the performance of the method on simulated and a real high-dimensional data set.

引用

页码：197 / 214

页数：18

共 33 条

[11]

Huang J, 2008, STAT SINICA, V18, P1603

[12] Fixed and Random Effects Selection in Mixed Effects Models [J].