期刊
STAT
卷 2, 期 1, 页码 86-101出版社
WILEY
DOI: 10.1002/sta4.20
关键词
group lasso; multiple functional predictors; penalized estimation; variable selection
资金
- NIH [R00ES017744]
- NCSU Faculty Research & Professional Development Fund [2012-2704, 2012-2673]
- NSF [DMS 1007466]
- NATIONAL INSTITUTE OF ENVIRONMENTAL HEALTH SCIENCES [R00ES017744] Funding Source: NIH RePORTER
Modern research data, where a large number of functional predictors is collected on few subjects are becoming increasingly common. In this paper we propose a variable selection technique, when the predictors are functional and the response is scalar. Our approach is based on adopting a generalized functional linear model framework and using a penalized likelihood method that simultaneously controls the sparsity of the model and the smoothness of the corresponding coefficient functions by adequate penalization. The methodology is characterized by high predictive accuracy, and yields interpretable models, while retaining computational efficiency. The proposed method is investigated numerically in finite samples, and applied to a diffusion tensor imaging tractography data set and a chemometric data set. Copyright (C) 2013 John Wiley & Sons, Ltd.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据