Sequential Co-Sparse Factor Regression.

Sequential Co-Sparse Factor Regression.
复制标题

DOI:
10.1080/10618600.2017.1340891
复制
发表时间:
2017
期刊:
Journal of computational and graphical statistics : a joint publication of American Statistical Association, Institute of Mathematical Statistics, Interface Foundation of North America
影响因子:
--
通讯作者:
Chen K
Chen K
中科院分区:
其他
文献类型:
--
作者:
Mishra A;Dey DK;Chen K

文献摘要

参考文献

被引文献

相似文献

在多元回归模型中,对回归分量矩阵进行稀疏奇异值分解可以降低模型的维数,便于模型的解释。然而,这种分解的恢复仍然非常具有挑战性,主要是由于同时存在正交约束和余稀疏正则化。通过深入研究底层的统计数据生成机制,我们重新制定的问题作为一个有监督的稀疏因子分析,并开发了一个有效的计算程序,命名为顺序因子提取通过稀疏单位秩估计(SeCURE),完全绕过正交性要求。在每一步,问题减少到一个稀疏的多元回归与单位秩约束。很好地,每个顺序提取的稀疏和单位秩系数矩阵自动导致其对奇异向量的共稀疏性。因此,每个潜在因子是预测因子的稀疏线性组合,并且可能仅影响响应的子集。所提出的算法保证收敛,并确保有效的计算,即使不完整的数据和/或强制执行精确的正交性是必要的。我们的估计享受甲骨文性质渐近的非渐近误差界进一步揭示了一些有趣的有限样本的估计行为。SeCURE的功效通过模拟研究和遗传学中的两个应用程序来证明。
In multivariate regression models, a sparse singular value decomposition of the regression component matrix is appealing for reducing dimensionality and facilitating interpretation. However, the recovery of such a decomposition remains very challenging, largely due to the simultaneous presence of orthogonality constraints and co-sparsity regularization. By delving into the underlying statistical data generation mechanism, we reformulate the problem as a supervised co-sparse factor analysis, and develop an efficient computational procedure, named sequential factor extraction via co-sparse unit-rank estimation (SeCURE), that completely bypasses the orthogonality requirements. At each step, the problem reduces to a sparse multivariate regression with a unit-rank constraint. Nicely, each sequentially extracted sparse and unit-rank coefficient matrix automatically leads to co-sparsity in its pair of singular vectors. Each latent factor is thus a sparse linear combination of the predictors and may influence only a subset of responses. The proposed algorithm is guaranteed to converge, and it ensures efficient computation even with incomplete data and/or when enforcing exact orthogonality is desired. Our estimators enjoy the oracle properties asymptotically; a non-asymptotic error bound further reveals some interesting finite-sample behaviors of the estimators. The efficacy of SeCURE is demonstrated by simulation studies and two applications in genetics.
DOI: 10.1214/11-aos876
发表时间: 2011-04-01
影响因子: 4.5
作者:
Bunea, Florentina;She, Yiyuan;Wegkamp, Marten H.
通讯作者: Wegkamp, Marten H.
DOI: 10.1007/s10208-009-9045-5
发表时间: 2009-12-01
影响因子: 3
作者:
Candes, Emmanuel J.;Recht, Benjamin
通讯作者: Recht, Benjamin
DOI: 10.1091/mbc.9.12.3273
发表时间: 1998-12-01
影响因子: 3.3
作者:
Spellman, PT;Sherlock, G;Futcher, B
通讯作者: Futcher, B
DOI: 10.1007/s00186-007-0161-1
发表时间: 2007-12-01
影响因子: 1.2
作者:
Gorski, Jochen;Pfeuffer, Frank;Klamroth, Kathrin
通讯作者: Klamroth, Kathrin
DOI: 10.1126/science.1075090
发表时间: 2002-10-25
期刊: SCIENCE
影响因子: 56.9
作者:
Lee, TI;Rinaldi, NJ;Young, RA
通讯作者: Young, RA