期刊
JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS
卷 19, 期 4, 页码 947-962出版社
AMER STATISTICAL ASSOC
DOI: 10.1198/jcgs.2010.09188
关键词
High dimension low sample size; Lasso; Multiple output regression; Sparsity
资金
- Yahoo
- National Science Foundation [DMS-0805798, DMS-0705532, DMS-0748389]
- Direct For Mathematical & Physical Scien
- Division Of Mathematical Sciences [0805798] Funding Source: National Science Foundation
We propose a procedure for constructing a sparse estimator of a multivariate regression coefficient matrix that accounts for correlation of the response variables. This method, which we call multivariate regression with covariance estimation (MRCE), involves penalized likelihood with simultaneous estimation of the regression coefficients and the covariance structure. An efficient optimization algorithm and a fast approximation are developed for computing MRCE. Using simulation studies, we show that the proposed method outperforms relevant competitors when the responses are highly correlated. We also apply the new method to a finance example on predicting asset returns. An R-package containing this dataset and code for computing MRCE and its approximation are available online.
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